1. health and fitness
2. disease
3. diabetes

Faculty of Life Sciences
Department of Health Sciences
Evaluation of the Effectiveness of Lifestyle Interventions in the
primary prevention of diabetes mellitus type II in high-risk
individuals
Master thesis
for the degree
Master of Public Health
Date of submission: 20.12.2013
Submitted by:
Karin Riemann-Lorenz
Matriculation No. 2042273
First Examiner: Prof. Dr. Joachim Westenhöfer (HAW)
Second Examiner: Dr. Klaus Koch (IQWIG)
Abstract
Background/research question: The increasing prevalence of type II diabetes
mellitus (T2DM) raises the question as to whether lifestyle interventions
(diet/exercise) can delay or prevent the manifestation of T2DM and reduce the risk
of micro- and macrovascular complications and premature death in high-risk
individuals. The primary aim of this thesis was to determine whether systematic
reviews exist that are able to answer the research question and thus provide a
basis for the production of evidence-based health information (EBHI). Secondly,
evidence on patient relevant outcomes like cardiovascular disease and mortality
on the basis of individual RCTs and epidemiological studies was to be collected.
Methods: PubMed, the Cochrane Library and DARE have systematically been
searched for systematic reviews of controlled trials (CTs) in June 2013. The
DIMDI, NICE and AHRQ websites were manually searched. Systematic reviews
were included if they met predefined inclusion criteria and achieved an Oxman &
Gyuatt Index ≥ 5. Two researchers independently screened 710 titles/abstracts
and the resulting 48 full-text articles.
Results: 9 relevant systematic reviews have been identified, of which 4 had the
required quality. They differed in terms of the populations, interventions and outcomes included, as well as the search date. The most appropriate review included
11 randomized CTs in a meta-analysis: The pooled hazard ratio versus controls
was 0.51 (95% CI 0.43-0.62). Assuming an annual diabetes incidence of 11% as in
the US Diabetes Prevention Program the calculated absolute risk of diabetes in the
intervention groups would be 5.61% (95% CI 4.73- 6.82). Significant reductions in
long term adverse health outcomes like CVD risk, CVD or all-cause mortality could
neither be found in the 4 identified reviews nor in the individual RCTs. However,
long term follow-up data of individual RCTs indicated benefits of lifestyle
interventions in terms of microvascular complications like diabetic retinopathy and
HRQoL.
Conclusions: Lifestyle interventions can prevent or delay the diagnosis of T2DM
in high-risk individuals. The findings translate into an estimated effect of about 5
out of 100 individuals by reducing an assumed annual diabetes progression rate of
11 out of 100 to about 6 out of 100 individuals. The systematic procedure to find
the best available external evidence resulted in a suitable basis for the production
of EBHI on diabetes prevention in high-risk individuals. Whether the applied
evaluation method is the most suitable for creating EBHI is worth consideration.
i
Acknowledgements
I would like to express my gratitude to my family, my friends and colleagues, who
have always encouraged and supported me during my studies and during the time
in which I wrote this master thesis. My gratitude goes to my first supervisor Prof.
Joachim Westenhöfer and my second supervisor Dr. Klaus Koch, who was very
open minded when I suggested to write my thesis on diabetes prevention in
cooperation with the IQWIG. My special thanks go to Dr. Martina Ehrlich for her
very friendly and cooperative way of working together in this project and to my
friend Dr. Kerstin Meyer, who gave me helpful feedback on English academic
writing.
ii
Content
1
Introduction .......................................................................................................... 1
2
Research Question and Objectives ...................................................................... 2
3
Theoretical Background ....................................................................................... 3
3.1
Diabetes mellitus type 2 ....................................................................................... 3
3.1.1
Definition and diagnostic criteria ............................................................... 3
3.1.2
Risk Factors ............................................................................................. 5
3.1.3
Definition of high-risk individuals .............................................................. 6
3.1.4
Prevalence of diabetes mellitus type 2 (T2DM) ........................................ 7
3.1.5
Health risks resulting from hyperglycemia and T2DM ............................. 10
3.1.6
Impact on Health Related Quality of Life (HRQoL) ................................. 12
3.2
The Concept of Health Literacy and its implications for Evidence Based Health
Information (EBHI) ............................................................................................. 13
3.3
The Institute for Quality and Efficiency in Health Care (IQWIG) ......................... 16
3.3.1
Legal basis and responsibilities .............................................................. 16
3.3.2
IQWIG Health Information ...................................................................... 18
4
3.3.2.1
Objectives and Characteristics........................................................... 18
3.3.2.2
Patient-centered communication ........................................................ 19
3.3.2.3
Patient relevant outcomes ................................................................. 20
3.3.2.4
Method of information retrieval: systematic literature research,
screening and quality assessment ..................................................... 22
Methods ............................................................................................................. 24
4.1
Literature Research ........................................................................................... 24
4.2
Screening, selection of reviews and grading with the Oxman & Guyatt Index .... 25
5
Results .............................................................................................................. 27
5.1
Identified Systematic Reviews ........................................................................... 27
5.1.1
Overall results of identified systematic reviews ....................................... 32
5.1.2
The ScHARR Review ............................................................................. 34
5.2
5.1.2.1
Methods ............................................................................................. 34
5.1.2.2
Included RCTs ................................................................................... 34
5.1.2.3
Meta-analysis on prevention of T2DM................................................ 55
5.1.2.4
Secondary Outcomes ........................................................................ 57
Patient-relevant outcomes ................................................................................. 57
5.2.1
CVD events, CVD mortality and all-cause mortality ................................ 57
iii
5.2.2
Microvascular complications ................................................................... 59
5.2.3
Health Related Quality of Life (HRQoL) .................................................. 59
5.3
Results of interest for EBHI................................................................................ 60
5.3.1
Adherence to lifestyle change and dose-response relationship .............. 60
5.3.2
Quality criteria of effective interventions ................................................. 62
5.3.3
Absolute Risk Reduction and long term progression rates to T2DM ....... 63
6
Discussion ......................................................................................................... 65
6.1
Comparison of results with previous reviews ..................................................... 65
6.2
Comparison with guidelines of relevant organizations ........................................ 66
6.3
Methodological considerations ........................................................................... 72
6.4
The relative importance of EBHI on diabetes prevention from a public health
perspective. ....................................................................................................... 75
7
Conclusions and Outlook ................................................................................... 78
References: ..................................................................................................................... 80
Statutory Declaration ....................................................................................................... 90
Appendix ......................................................................................................................... 91
iv
List of Abbreviations
ADA
American Diabetes Association
AHRQ
Agency for Healthcare Research and Quality
BÄK
Bundesärztekammer
BMG
Bundesministerium für Gesundheit
BMJ
Bundesministerium der Justiz
CVD
Cardiovascular Disease
CG
Control Group
DARE
Database of Abstracts of Reviews of Effects
DEGS
Studie zur Gesundheit Erwachsener in Deutschland
DPP
Diabetes Prevention Program
DPPOS
Diabetes Prevention Program Outcomes Study
DPPRG
Diabetes Prevention Program Research Group
DPS
Diabetes Prevention Study
EBM
Evidence Based Medicine
EBHI
Evidence Based Health Information
ERFC
Emerging Risk Factor Collaboration
FPG
Fasting Plasma Glucose
H(b)A1C
Glycated Hemoglobin
HR
Hazard Ratio
HRQoL
Health Related Quality of Life
IDF
International Diabetes Federation
IDPP
Indian Diabetes Prevention Program
IG
Intervention Group
IGT
Impaired Glucose Tolerance
IOM
Institute of Medicine
LI
Lifestyle Intervention
NICE
National Institute for Health and Clinical Excellence
NGT
Normal Glucose Tolerance
NNT
Number Needed to Treat
OGGT
Oral Glucose Tolerance Test
PICO
Population Intervention Control Outcome
RCT/RCTs
Randomized Controlled Trial/Randomized Controlled Trials
RKI
Robert Koch Institut
SCB
Social Code Book
SHI
Statutory Health Insurance
ST2DM
Screen detected type 2 diabetes mellitus
T2DM
Diabetes mellitus type 2
WHO
World Health Organization
YLL
Years of Life Lost
v
List of Tables
Table 1:
Currently recommended diagnostic criteria for diabetes mellitus type 2
(T2DM), Impaired Glucose Tolerance (IGT) and Impaired Fasting Glucose
(IFG) by WHO, ADA and BÄK
Table 2:
Modifiable and non-modifiable risk factors for T2DM
Table 3:
Categories for increased risk for diabetes
Table 4:
Lifetime prevalence of known diabetes – data from the DEGS
Table 5:
Lifetime prevalence of known diabetes according to gender and
socioeconomic status– data from the DEGS
Table 6:
Prevalence of diagnosed diabetes in the population aged 18 years and
older
Table 7:
Key requirements for evidence based health information (EBHI) according
to “Gute Praxis Gesundheitsinformation”
Table 8:
Characteristics of EBHI according to the IQWIG Handbook of “General
Methods” version 4.0 of 23.09.2011
Table 9:
Patient-relevant and surrogate endpoints in diabetes
Table 10:
PICO scheme for the literature research
Table 11:
Main characteristics and results of the systematic reviews included
Table 12:
Basic characteristics of included RCTs
Table 13:
Progression to diabetes and regression to NGT in larger trials
Table 14:
Overview of recommendations from ADA, IMAGE Study Group and NICE
List of Figures
Figure 1:
Estimated numbers of YLL owing to diabetes
Figure 2:
Flow chart of screening process
Figure 3:
Meta-analysis of lifestyle interventions
Figure 4:
Cumulative incidence of diabetes in the DPP from randomization to year 10
all participants
Figure 5:
The main determinants of health
vi
1
Introduction
Diabetes mellitus type 2 is a disease of major public health concern in high-income
countries and low- and middle-income countries alike. According to the
International Diabetes Federation (IDF) the number of people with diabetes is
increasing in every country (IDF, 2012). In the European region prevalence rates
are rising among all ages mostly due to increases in overweight and obesity,
unhealthy diet and physical inactivity (WHO EUROPE, 2013). It is estimated that
about 60 million people in the European Region live with diabetes, 10.3 % of men
and 9.6 % of women aged 25 and over with substantial differences between
individual countries (WHO EUROPE, 2013; Thelen et al., 2013). For Germany
results of the German Health Interview and Examination Survey (DEGS) indicate
that at least 4.6 million Germans aged 18 to 79 years have been diagnosed with
diabetes at some point in life, which would translate to a lifetime prevalence rate of
7.2% (Heidemann et al., 2013).
There are two main reasons why preventive strategies are urgently needed
(Lindström et al., 2006) to limit the burden of disease resulting from T2DM and its
complications for patients and health care systems: population aging and decrease
of the age of onset of T2DM.
As “Age is one of the strongest risk factors for T2DM.” (Paulweber et al., 2010, p.
S4) the number of individuals affected will increase in an aging society as found in
Germany. In the DEGS diabetes prevalence rates ranged from under 5% among
those under 50 years of age to more than 20% of those being older than 70
(Heidemann et al., 2013). Furthermore, there is evidence that the age of onset
decreases in countries with increasing obesity prevalence (Paulweber et al., 2010).
For Germany an increase in diabetes prevalence in the age group of 25 to 69 as
well as in the adult population over the age of 18 in the last decade has been
verified (Heidemann et al., 2011).
Looking from the patients` perspective, surveys have shown that an overwhelming
majority is interested to learn what they themselves can do to maintain good health
or reduce disease consequences. 89 % expected advice from their general practitioner on how to reduce the risk of future illness and 85% wanted to know how they
could stay healthy in the future (Little et al., 2001). As time available to the individual patient is usually very limited when visiting a doctor, evidence based health information (EBHI) on the effectiveness of lifestyle changes in the prevention of
T2DM could be very useful for patients, consumers and doctors alike.
1
2
Research Question and Objectives
As pointed out EBHI on the effectiveness of diabetes prevention could be one
component of a strategy to meet the information needs of consumers and
patients and at the same time promote the efforts towards preventive
strategies. In Germany the Institute for Quality and Efficiency in Health Care
(IQWIG) is legally obliged to provide EBHI for consumers and patients on
diseases of substantial epidemiological relevance (IQWIG, 2011). As a prerequisite evidence has to be assessed according to the methods of evidence based
medicine (EBM). The goal of this master thesis was therefore to evaluate the
effectiveness of lifestyle interventions to prevent or delay diagnosis of T2DM in
high-risk individuals applying the methods of EBM. The result of this evaluation
aims to build the basis for the production of EBHI on the topic which is to be
published on the IQWIG website in the near future.
Apart from evaluating the possibility of delaying or preventing T2DM, an
additional objective of this thesis was to search for evidence for the benefit of
LI in terms of patient relevant outcomes like CVD risk, CVD mortality, all-cause
mortality, morbidity or HRQoL. Deviating from the IQWIG`s usual practice at
the time of this research to use primarily systematic reviews of CT`s as a basis
for EBHI, also insights that have been gained from individual RCTs or epidemiological studies will be presented in this master thesis.
Finally, using the example of prevention of T2DM, this bit of research considers
the question, whether or not the search for systematic reviews of controlled
trials may be an appropriate method for the assessment of evidence in primary
prevention and the preparation of EBHI.
2
3
Theoretical Background
3.1
Diabetes mellitus type 2
3.1.1 Definition and diagnostic criteria
Diabetes mellitus type 2 (T2DM) is defined as a metabolic disorder that is primarily
characterized by chronic hyperglycemia induced by disturbances in insulin
secretion, insulin action or both (WHO, 1999; BÄK et al., 2013). At present there is
no unique biological marker that distinguishes people with diabetes from nondiabetic but hyperglycemic individuals. Therefore plasma glucose levels remain the
basis for the diagnosis of T2DM and the definition of intermediate hyperglycemia,
namely Impaired Glucose Tolerance (IGT) and Impaired Fasting Glucose (IFG)
(ADA, 2003; WHO, 2006). In a report of a WHO/IDF Consultation in 2006 the
WHO stated: “In the absence of a more specific biological marker to define
diabetes, plasma glucose estimation remains the basis of diagnostic criteria”
(WHO, 2006, p.9). However, the exact cut-off points that distinguish hyperglycemia
from diabetes mellitus type 2 have changed over time and still are a matter of
debate (WHO, 2006).
The different, currently recommended diagnostic criteria for T2DM, IGT and IFG
are summarized in table 1.
Especially the definition of IFG is contentious. The WHO pointed out, that lowering
the cut-point for IFG as proposed by the ADA in 2003 would lead to a significant
increase in IFG prevalence with enormous impact on individuals and health
systems (WHO, 2006). Moreover, the risk of progressing to T2DM is much higher
in people with FPG levels of > 6.1 mmol/l than in those with FPG levels of 5.6-6.0
mmol/l (Gillett et al., 2012). In addition the WHO stated that there is a lack of
evidence of any benefit of lowering the IFG threshold with regard to progression to
diabetes or adverse clinical outcomes (WHO, 2006).
3
Table 1: Currently recommended diagnostic criteria for diabetes mellitus type 2 (T2DM), Impaired Glucose Tolerance (IGT) and Impaired
Fasting Glucose (IFG) by WHO, ADA and BÄK (WHO, 2006; ADA, 2003; ADA, 2013; BÄK et al., 2013).
Diagnostic Criteria of
Diabetes
Fasting plasma glucose
2-h plasma glucose
IGT
Fasting plasma glucose
WHO, 2006 and WHO, 2011
ADA, 2003 and 2013
BÄK et al., 2013
≥ 7,0 mmol/ (126 mg/dl)
≥ 7,0 mmol/ (126 mg/dl)
> 7,0 mmol/ (126 mg/dl)
or
or
and/or
≥ 11,1 mmol/l (200 mg/dl)
≥ 11,1 mmol/l (200 mg/dl)
≥ 11,1 mmol/l (200 mg/dl)
or
or
and/or
HbA1c levels of 48 mmol/mol
(6.5%)
A1C ≥ 6,5%
HbA1c ≥ 48 mmol/mol (≥ 6.5%)
< 7.0 mmol/l (126 mg/dl)
-
-
7,8 – 11 mmol/l
(140 –199 mg/dl)
≥ 7.8 and < 11.1 mmol/l (140
mg/dl and 200 mg/dl)
5.6–6.9 mmol/l
(100–125 mg/dl)
≥ 5.6 mmol/ and < 7.0 mmol/l
(≥ 100 mg/dl and < 126 mg/dl)
and
2-h plasma glucose
IFG
Fasting plasma glucose
≥ 7.8 and < 11.1 mmol/l (140
mg/dl and 200 mg/dl)
6.1 to 6.9 mmol/l
(110 mg/dl to 125 mg/dl)
and (if measured)
2-h plasma glucose
< 7.8 mmol/l (140 mg/dl)
4
3.1.2 Risk Factors
T2DM is based on a genetically determined, multi-factorial predisposition. The
clinical manifestation occurs under the influence of modifiable and non-modifiable
risk factors, which are summarized in table 2 (BÄK et al., 2013, WHO, 1999,
Paulweber et al., 2010).
Table 2: Modifiable and non-modifiable risk factors for T2DM (Paulweber et al.,
2010, p. S5)
The degree of risk for the individual of developing T2DM is principally determined
by the number and severity of risk factors.
Gillett et al. emphasized the importance of overweight and obesity as a risk factor
for developing T2DM (Gillett et al., 2012). Especially central adiposity – measured
by waist circumference – has been shown to be a predictor of risk for developing
T2DM (Diabetes Prevention Program Research Group, 2006). Moreover, central
adiposity is a particular risk even in people with normal BMI (Han et al., 2006).
In a Canadian study Hart et al. investigated the relationship between BMI in middle
age and risk of T2DM using data from large prospective studies. Compared with
the normal weight group the odds ratios for incident diabetes in men, adjusted for
age, social class, smoking and systolic blood pressure, were 2.56 (95% CI 1.91–
3.42) and 6.48 (95% CI 4.65–9.03) in the overweight and the obese group,
respectively. In women the numbers were 2.54 (95% CI 1.94–3.32) and 5.43 (95%
CI 4.07–7.26), respectively (Hart et al., 2007).
Mozaffarian et al. investigated the impact of a combination of five lifestyle factors
on the incidence of new-onset T2DM in an older general population (Mozaffarian et
al., 2009). Included lifestyle factors were physical activity level, dietary score
5
composed of fibre intake, fat quality and mean glycemic index, smoking status,
alcohol use and body weight (BMI and measures of waist circumference). They
found that “..after adjustment for age, sex, race, educational level, annual income
and other lifestyle factors simultaneously, each lifestyle risk factor was
independently associated with incidence of diabetes, with 26%, 31%, 23%, 34%,
45%, and 46% lower risk among older adults in the low-risk groups for physical
activity level, dietary habits, smoking habits, alcohol use, BMI, and waist
circumference, respectively.” (Mozaffarian et al., 2009, p. 801). Participants who
were in the low risk group for several lifestyle factors (physical activity, dietary
score, smoking and alcohol habits) reduced their risk for incident diabetes by 82%
(relative risk 0,18; 95% CI 0.06- 0.56) compared to all other participants.
3.1.3 Definition of high-risk individuals
In the literature no generally acknowledged definition of individuals at high risk for
developing T2DM could be found. However, people with IFG and/or IGT are
usually considered to be at high risk although the progression rates from IFG and
IGT to T2DM differ between studies in different populations (Gillett et al., 2012).
Age, baseline levels of FPG, 2-hour glucose levels, HbA1c, grade of central
adiposity and BMI have been found to be predictors of progression (Gillett et al.,
2012). A report issued by the Agency for Healthcare Research and Quality
(AHRQ) in 2005 found consistent evidence that IFG and IGT are both indicators for
an increased risk for developing T2DM and calculated pooled relative risks. In
people with IGT it was 6.02 (95% CI 4.66 to 7.38), in people with IFG it was 4.70
(95% CI 2.71 to 6.70) and in people with IFG and IGT it was 12.21 (95% CI 4.32 to
20.10) compared to the NGT groups (Santaguida et al., 2005). In addition the
authors calculated estimates for the attributable risk in the exposed groups
(exposure being IGT, IFG or combined IGT&IFG) over the entire study duration in
more than 30 studies for the progression to T2DM. The numbers reported were AR
for the IGT group: 52.8% to 97.0%, for the IFG group: 57.3% to 86.9% and for the
combined IGT & IFG group: 78.6% to 93.0%. Thus the authors concluded that …”if
there is a causal relationship between IFG or IGT and progression to DM, as many
as 97% of cases of DM within the IGT group could be prevented by treating or
eliminating dysglycemia.” (Santaguida et al., 2005, p. 30).
In its position statement on the diagnosis and classification of diabetes mellitus
from 2013 the ADA summarized categories of increased risk for diabetes as shown
in table 3, emphasizing that the evaluation of a patient’s risk should also
6
incorporate a global risk factor assessment for diabetes and cardiovascular
disease (ADA, 2013).
Table 3: Categories for increased risk for diabetes* (ADA, 2013, p. S13)
FPG 100 mg/dl (5.6mmol/l) to 125mg/dl (6.9 mmol/l) [IFG] or
2-h PG in the 75-g OGTT 140 mg/dl (7.8mmol/l) to 199 mg/dl (11.0 mmol/l)
[IGT] or
A1C 5.7–6.4%
*For all three tests, risk is continuous, extending below the lower limit of the range and becoming
disproportionately greater at higher ends of the range.
Besides IGT and IFG some organizations also consider metabolic syndrome as a
risk
factor
for
the development
of
T2DM.
In
the
German „Nationale
VersorgungsLeitlinie Therapie des Typ-2-Diabetes“ the organizations involved
stated that the clinical manifestation of T2DM often occurs under the influence of
risk factors which present in the form of the metabolic syndrome (BÄK et al.,
2013). According to the authors the main characteristics of metabolic syndrome
are: abdominal obesity (waist circumference in men > 94 cm, in women > 80 cm),
insulin resistance, hyperinsulinemia, impaired glucose tolerance, dyslipidemia,
albuminuria and hypertension (own translation from BÄK et al., 2013).
3.1.4 Prevalence of diabetes mellitus type 2 (T2DM)
According to the IDF the number of people with diabetes is increasing in every
country. IDF estimated that in 2012 more than 371 million people suffered from
diabetes worldwide (IDF, 2012). Almost 80 % of diabetes related deaths – 3.4
million annually - occur in low- and middle-income countries (WHO EUROPE,
2013). In the European Region prevalence rates are increasing among all ages
mostly because of increases in overweight and obesity, unhealthy diet and
physical inactivity (WHO EUROPE, 2013). It is estimated that about 60 million
people in the European Region live with diabetes, 10.3 % of men and 9.6 % of
women aged 25 and over with substantial differences in the individual countries
(WHO EUROPE, 2013; Thelen et al., 2013).
As there is no reporting obligation or central register for T2DM in Germany
prevalence estimates have mostly been based on data from regional studies,
health insurance companies or general practices assessed in the past. Up to date
nationwide and population based data has only recently been provided by the
telephone surveys “German Health Update” (GEDA 2009 and GEDA 2010)
7
(Heidemann et al., 2013). Data from GEDA 2009 showed a self-reported lifetime
prevalence of diabetes in a population aged 18 and older and living in private
homes of 8.8% (9.3 % in women and 8.2% in men). An extrapolation of these
results to the adult population in Germany would amount to 5.98 million individuals
having encountered diabetes over their lifetimes.
Current prevalence data have also recently been published on the basis of the
German Health Interview and Examination Survey (DEGS), which has been
conducted from 2008 to 2011 by the Robert Koch Institute (RKI) in Berlin
(Heidemann et al., 2013). In a representative sample of the German population
aged 18 to 79 in total 591 of the 7080 participants stated that they had been
diagnosed with diabetes by a doctor at some point in life. Among the 591 cases
were 8 with diabetes type 1 and 42 with gestational diabetes, which means that
91.5 % of reported diabetes cases were T2DM. Lifetime prevalence for different
age groups stratified for gender is shown in table 4.
Table 4: Lifetime prevalence of known diabetes – data from the DEGS
(Heidemann et al., 2013)
According to the results of DEGS lifetime prevalence of known diabetes is 7.2%
(7.4 % for women and 7.0% for men). The results indicate that at least 4.6 million
Germans aged 18 to 79 years have been diagnosed with diabetes at some point in
life. For both sexes diabetes prevalence increases with age, rising from under 5%
among those under 50 years of age to more than 20% of those being older than 70
(Heidemann et al., 2013). An increase in diabetes prevalence of comparable size
with age has also been found in the GEDA 2009 (Heidemann et al., 2011).
8
Prevalence of diabetes was also higher in those with low socioeconomic status,
especially in women as table 5 shows.
Table 5: Lifetime prevalence of known diabetes according to gender and
socioeconomic status– data from the DEGS (Heidemann et al., 2013)
When compared with data from the German National Health Interview and
Examination Survey (GHNIES) from 1998, the DEGS study found an absolute
increase in diabetes prevalence in the last decade of 2% (lifetime prevalence of
5.2% in GNHIES 98 and 7.2% in DEGS). After taking demographic aging into
account a significant increase of 1.4% remains, meaning that only 0.6 % of the
increase can be attributed to population aging (Heidemann et al., 2013). An
increase in diabetes prevalence in the age group of 25 to 69 as well as in the adult
population over the age of 18 in the last decade has also been verified by GEDA
2009 as shown in table 6 (Heidemann et al., 2011).
Table 6: Prevalence of diagnosed diabetes in the population aged 18 years and
older (Heidemann et al., 2011).
9
3.1.5 Health risks resulting from hyperglycemia and T2DM
Hyperglycemia or T2DM can exist in an individual over a long period of time
without symptoms, nevertheless these conditions are associated with substantial
health risks (RKI, 2005; WHO, 1999).
In the Australian Diabetes, Obesity, and Lifestyle Study (AusDiab) 10428
participants were followed up for a median of 5.2 years. After adjustment for well
known CVD risk factors, individuals with known diabetes at baseline had an allcause mortality risk that was 2 times greater than those with normal glucose
tolerance (Barr et al., 2007). These findings have been confirmed in 2011 by a
meta-analysis of 97 prospective studies with data on 123,205 deaths among
820,900 people (Emerging Risk Factors Collaboration (ERFC), 2011). The ERFC
calculated a 1.80 (95% CI 1.71 to 1.90) HR for death from any cause among
persons with diabetes compared to non-diabetics and a 2.32 HR (95% CI 2.11 to
2.56) for death from vascular causes, respectively. The authors estimated that a
50-year-old person with diabetes died on average 6 years earlier than a person
without diabetes. About 40% of excess deaths in diabetes patients were
attributable to nonvascular deaths (ERFC, 2011). Figure 1 shows the estimated
future Years of Life Lost (YLL) owing to diabetes depending on the age of onset.
Figure 1: Estimated numbers of YLL owing to diabetes (ERFC, 2011, p. 838)
However, not only diabetes mellitus increases mortality rates but also milder forms
of impaired glucose metabolism. As early as in 1999 Coutinho et al. found a
progressive relationship between fasting glucose levels and cardiovascular risk for
glucose levels well below the diabetes threshold in a systematic overview and
meta-regression analysis of cohort studies of non-diabetic individuals (Coutinho et
10
al, 1999). “…compared with the reference fasting glucose of 4.2 mmol/l (75 mg/dl),
a fasting glucose of 6.1 mmol/l (110 mg/dl, the threshold value for the classification
of impaired fasting glucose [10]) was associated with a relative risk of
cardiovascular events of 1.33 (95% CI 1.06–1.67); a 2-h glucose of 7.8 mmol/l
(140 mg/dl, the threshold value for impaired glucose tolerance) was associated
with a relative risk of cardiovascular events of 1.58 (95% CI 1.19–2.10).” (Coutinho
et al, 1999, p. 237).
Mortality from all causes was also increased in those with IGT and IFG at baseline
in the AusDiab study but to a lesser extent than those with known T2DM at
baseline. Also CVD mortality was significantly higher in those with T2DM and IFG
at baseline, but not in those with IGT (Barr et al., 2007). The authors concluded
that there is a “…strong association between abnormal glucose metabolism and
mortality, and it suggests that this condition contributes to a large number of CVD
deaths in the general population.” (Barr et al., 2007, p. 151). They suggest that
CVD prevention strategies should address not only people with T2DM but also
those with milder forms of hyperglycemia (IFG and IGT) (Barr et al., 2007).
Considering the threshold for IGT, other studies have shown that there is an
increase in risk for cardiovascular disease and mortality below the IGT treshold of
7.8 mmol/l (THE DECODE STUDY GROUP, 2003; Levitan et al., 2005). Also the
WHO after reviewing the literature came to the conclusion that “…the risk of future
diabetes, premature mortality and cardiovascular disease begins to increase at 2–
h plasma glucose levels below the IGT range.” (WHO, 2006, p. 19) Therefore the
WHO experts recommended to consider “…replacing this category of intermediate
hyperglycaemia by an overall risk assessment for diabetes, cardiovascular
disease, or both, which includes a measure of glucose as a continuous variable.“
(WHO, 2006, p. 19).
In summary the WHO experts concluded that…”there are an abundance of data
indicating that hyperglycemia is harmful. However there are limitations in the data
and the methodologies used to derive cut-points at which this level of harm is
specifically increased and which clearly differentiate diabetes from non-diabetes”
(WHO, 2006, p. 12). According to WHO the main problem is “…placing a specific
cutpoint on a continuous variable.” (WHO, 2006, p.13).
11
Besides increases in mortality risk long-term damage, dysfunction and functional
limitations of various organs - especially the eyes, kidneys, nerves and the
cardiovascular system - can be the result of chronic hyperglycemia in diabetes
(BÄK et al., 2013; RKI, 2005). Diabetes mellitus type 2 ”..is associated with
reduced life expectancy, significant morbidity due to specific diabetes related
microvascular complications, increased risk of macrovascular complications
(ischaemic heart disease, stroke and peripheral vascular disease) and diminished
quality of life” (WHO, 2006, p. 5).
3.1.6 Impact on Health Related Quality of Life (HRQoL)
Diabetes patients experience reduced HRQoL compared with people with no
chronic disease. The presence of diabetes-related complications worsens HRQoL
even more (Rubin and Peyrot, 1999). Reductions in HRQoL are associated with
difficulties in performing everyday tasks like walking, climbing stairs or bending and
reduced scores in “vitality” and “general heath” in the SF36 questionnaire (Tapp et
al., 2006). However, HRQoL can be reduced even before the diagnosis of diabetes
mellitus especially in those with IGT (Tapp et al., 2006). Tapp et al. found “…a
gradual decrease in quality of life across categories of glucose tolerance status” in
a cross-sectional study with 10334 participants in Australia, for whom data on the
SF-36 questionnaire were available (Tapp et al., 2006, p. 158). People with IGT
generally had a lower score in different dimensions of HRQoL compared to
individuals with NGT. However, compared to individuals with known T2DM the
mean quality of life scores for bodily pain, general health perception, physical
functioning, vitality and other dimensions were higher in the IGT group (Tapp et al.,
2006).
12
3.2
The Concept of Health Literacy and its implications for Evidence Based
Health Information (EBHI)
In conjunction with the increasing importance of EBM the concept of health literacy
on the patients` side has become of growing interest in the last decade. Health
literacy has been defined by the Institute of Medicine (IOM) in 2004 as follows:
“Health literacy is the degree to which individuals can obtain, process and
understand basic health information and services they need to make appropriate
health decisions.” (IOM, 2004, p. 1).
Apart from statistical literacy – that is the ability to understand the meaning of
numbers, proportions and probabilities – as a basic prerequisite, other social
competencies and skills are fundamental to the concept of health literacy (MironShatz et al., 2011). A broader definition published by the WHO in 2010 described
health literacy as “…the cognitive and social skills which determine the motivation
and ability of individuals to gain access to, understand, and use information in
ways which promote and maintain good health. Health Literacy means more than
being able to read pamphlets and successfully make appointments. By improving
people's access to health information and their capacity to use it effectively, health
literacy is critical to empowerment. Defined this way, Health Literacy goes beyond
a narrow concept of health education and individual behaviour-oriented
communication, and addresses the environmental, political and social factors that
determine health.” (WHO, 2010)
Among the cognitive and social skills mentioned in the definition are a basic
understanding of scientific concepts, the knowledge about the inherent element of
uncertainty in scientific findings, the ability to judge the information source as
credible or not and the understanding, that health information is interpreted
differently depending on the beliefs, customs and social norms of different subgroups of society (Miron-Shatz et al., 2011).
However, health literacy in this comprehensive sense is rather the exception than
the rule among citizens or patients (see Gaissmaier & Gigerenzer, 2011 and
Miron-Shatz et al., 2011). Advocates for the improvement of health literacy expect
benefits in two areas: ethics and economics. The ethical aspect of the discussion
about patient rights and health literacy has been reflected in the German
legislation. According to German law citizens and patients have the right to
comprehensive information concerning their health and disease status as well as
13
understandable communication of the information (BMG, 2003). In February 2013
these rights have been summarized in a dedicated patient rights law (BMJ, 2013).
Closely linked with the ethical imperative of patient information are economic
considerations. Miron-Shatz et al. pointed out: “Since the Age of Enlightenment,
efforts have focused on educating citizens for their personal and the greater
societal good. ..., efforts to increase health literacy have been advocated as a
necessary condition for a better educated population – one capable of making
appropriate and informed health decisions and engaging in recommended health
behaviors.” (Miron-Shatz et al., 2011, p. 209) These considerations are supported
by findings that show “… that individuals with limited health literacy incur up to four
times greater cost in unnecessary doctor visits and hospital care, compared to
individuals with average health literacy. Increasing health literacy has been
associated with a number of positive outcomes: improved decision making, better
understanding of disease and treatment regimens, and adherence to prescribed
treatment options. It is thus possible that increased health literacy could translate
to lower costs…” (Miron-Shatz et al., 2011, pp. 209-210). Also the IOM stated in its
report from 2004: “Although causal relationships between limited health literacy
and health outcomes are not yet established, cumulative and consistent findings
suggest such a causal connection….Studies have shown that people with low
health literacy understand health information less well, get less preventive health
care—such as screenings for cancer—and use expensive health services such as
emergency department care more frequently.” (IOM, 2004, p. 1)
The comprehensive concept of health literacy outlined here also implies some
basic requirements for EBHI, which have been summarized in the German
publication “Gute Praxis Gesundheitsinformation” by Klemperer et al., 2010. The
key issues are listed in table 7.
14
Table 7: Key requirements for evidence based health information (EBHI) according
to “Gute Praxis Gesundheitsinformation” (own translation) (Klemperer et al., 2010)
•
The content must be based on the best available evidence.
•
EBHI rely on systematic search, selection, critical appraisal and review of
the existing literature with the aim to reduce bias and to take into account
the reliability of the results.
•
The absence of sufficiently strong evidence should be mentioned.
•
The information must be relevant for the target group and
comprehensible.
•
Citizens, healthcare users or appropriate organizations should be
involved.
•
Information on diseases should draw a realistic picture of the knowledge,
the frontiers of knowledge, the causes, the diagnosis and the progression
of the disease, coping strategies as well as the existing prevention, early
detection and treatment options.
•
Information on treatment results should focus on patient relevant
outcomes (e.g. mortality, morbidity, health related quality of life).
•
For the individual harm-benefit-assessment comparison of different
treatment options are sensible including the no treatment option.
•
Health information should be formulated in a non-directive way to allow
the user to decide in accordance with his or her own values and
preferences.
•
The numerical representation of probabilities is useful if reliable data are
available. For the representation of probabilities the absolute change in
risk is of primary importance.
For the communication of risks the following recommendations are made:
•
Starting point of the information should be the natural course of disease. This
includes the probability that the clinical picture improves, worsens or remains
constant without intervention and how often undesired outcomes occur.
•
Changes in risks should be communicated as absolute risk reduction.
•
The exclusive representation of relative risk reduction is unsuitable, because
large effects cannot be distinguished from small effects.
15
•
Different framing of identical numbers can lead to differences in the perception
of risks and the motivation of patients.
•
It might be appropriate to combine different representations of risks e.g.
absolute risk, relative risk, number needed to treat, graphs or comparisons
with everyday risks. (Klemperer et al., 2010, p. 67, own translation).
3.3
The Institute for Quality and Efficiency in Health Care (IQWIG)
3.3.1 Legal basis and responsibilities
The IQWIG was founded within the German Health Care Reform of 2004 as an
institution which is supposed to improve quality and efficiency in health care within
the German statutory health insurance. Its legal basis has been anchored in Social
Code Book V, its responsibilities are described in detail in § 139a of SCB V and
can be summarized as follows:
•
“Search for, assessment and presentation of current scientific evidence on
diagnostic and therapeutic procedures for selected diseases;
•
Preparation of scientific reports, expert opinions, and comments on quality and
efficiency issues of SHI services, taking age, gender, and personal
circumstances into account;
•
Appraisal of evidence-based clinical practice guidelines (CPGs) on the most
relevant diseases from an epidemiological point of view;
•
Issue of recommendations on disease management programmes (DMPs);
•
Assessment of the benefit and cost of drugs;
•
Provision of easily understandable information for all patients and
consumers on the quality and efficiency of health care services, as well
as on the diagnosis and treatment of diseases of substantial
epidemiological relevance.”
(IQWIG, 2011 p. 1; emphasis added)
As pointed out in the Institute`s method paper, the IQWIG is legally obliged by §
139a (4) SCB V to conduct its assessments on the basis of the internationally
agreed upon standards of EBM (IQWIG, 2011). The concept of EBM has been
defined by Sackett et al. in 1996 as follows: "Evidence-based medicine is the
conscientious, explicit and judicious use of current best evidence in making
decisions about the care of individual patients. The practice of evidence based
16
medicine means integrating individual clinical expertise with the best available
external clinical evidence from systematic research." (Sackett et al., 1996).
However, as the “best available evidence is often incomplete or unreliable”
(IQWIG, 2011, p 4) assessments in accordance with EBM standards aim at
determining uncertainty and describing it in terms of evidence levels. This is
supposed to help clinicians and their patients to make informed decisions, also
taking into account their own personal values (IQWIG, 2011).
According to §139b SCB V only the Federal Joint Committee (Gemeinsamer
Bundesausschuss) or the Federal Ministry of Health are allowed to commission the
IQWIG. In addition and in accordance with the institute`s legal remit to provide
health information, the IQWIG itself can choose topics for or translate
commissioned reports into health information for consumers and patients (IQWIG,
2011).
As the need for health information is potentially limitless, prioritizing is necessary.
In line with the IQWIG`s responsibility to provide information about diseases with
substantial epidemiological relevance, diseases with a high burden of disease are
favored. These could be diseases with high mortality, incidence or prevalence,
utilization of health care services or high treatment costs. Also absence from work
due to illness and loss in quality of life of those affected by the disease could be
relevant criteria. Furthermore, the IQWIG tries to find out and take into account
what consumers and patients might be interested in by using different information
sources like surveys, qualitative research and topics suggested by self help groups
or users of the website. Priority is also given to questions for which evidence
based answers exist (IQWIG, 2011).
Health information for patients and consumers are provided by the IQWIG
Department of Health Information using a defined method of evidence retrieval and
different publishing formats.
17
3.3.2 IQWIG Health Information
3.3.2.1 Objectives and Characteristics
The IQWIG Department of Health Information aims to improve health and patient
autonomy by increasing health and scientific literacy. Published health information
is therefore meant to:
•
“Support active and informed decision-making about health issues;
•
Promote the critical use of health care services;
•
Improve understanding of physical, mental and emotional health;
•
Improve understanding of medical and scientific information, including the
concept of evidence-based medicine; and
•
Enable support of patients by family and friends.” (IQWIG, 2011, p. 69)
To reach these goals the IQWIG considers it necessary …”to to be a reliable,
trusted and patient-centred information provider” (IQWIG, 2011, p. 69) which aims
at integrating patient values in decision-making into EBHI. Consequently the
IQWIG gives a comprehensive definition of EBHI; the characteristics are
summarized in table 8.
Table 8: Characteristics of EBHI according to the IQWIG Handbook of “General
Methods” version 4.0 of 23.09.2011. (IQWIG, 2011, p. 69)
•
“The content is based on clear scientific evidence, particularly systematic
reviews;
•
The information is developed following systematic methods which aim to
minimize bias and maintain neutrality;
•
Evidence-based communication techniques are used to meet the goals
of informing, supporting and empowering users;
•
Uncertainties as well as the potential for benefit and harm are discussed;
•
Language and framing are neutral and non-directive, so that people can
make their decisions in accordance with their own values; and
•
The information is updated so that it remains evidence-based.“
As pointed out the IQWIG puts special emphasis on the requirement to only
communicate findings with clear scientific evidence. Therefore the institute`s health
18
information products rely to a large extent on results of systematic reviews, which
have to fulfill certain minimum quality requirements in order to minimize
methodological flaws (details see chapter 3.3.2.4).
3.3.2.2 Patient-centered communication
A key challenge for the IQWIG Department of Health Information is to provide
information, which is comprehensible for different target groups in the
population while at the same time remaining scientifically accurate and
objective. This challenge is increased by the fact, that (health) literacy levels
show a great variation throughout the potential readers of the information
(IQWIG, 2011).
However, the IQWIG “…aims at a readability below university level.” (IQWIG,
2011, p. 84) and uses test readers and reader ratings as a measure to assess
understandability (IQWIG, 2011).
In addition the IQWIG Department of Health Information is committed to a nonpaternalistic model of patient communication. Its goal is not only to inform patients,
but also strengthen them in their patients' autonomy. Key values the IQWIG tries to
take into account are therefore:
•
“Demonstrate sensitivity and respect for user knowledge, values and
concerns, autonomy, cultural differences as well as gender, age and disabilityrelated interests,
•
Maintain a patient-centred, non-judgmental, non-directive and neutral style of
language; and
•
Respect readers’ time.” (IQWIG, 2011, p. 86)
To address the different needs of consumers the IQWIG produces health
information in different formats: Fact sheets for an easy to understand, short
information, feature articles providing comprehensive information and research
summaries, which usually present the results of systematic reviews or larger
studies. Fact sheets and research summaries focus mainly on the effects of
treatments, diagnostic tests and self-management strategies, topics that are
known to be of major interest to most people seeking health information.
Preventive and health promotion measures like changes in diet and physical
activity are included into the information product if they are regarded relevant for
the individual topic (IQWIG, 2011).
19
In order to make the website more attractive to users, increase understanding of
medical issues and support self-management strategies, supplementary items like
graphics, short animated films, interactive quizzes, calculators, online polls and
patient stories round off the information offered by the IQWIG Department of
Health Information (IQWIG, 2011).
3.3.2.3 Patient relevant outcomes
The benefit assessment of medical interventions and treatments and the
communication of the results is a predominant duty of the IQWIG. The assessment
of benefit “…is based on the results of studies investigating the effects of an
intervention on patient-relevant outcomes.” (IQWIG, 2011, p. 28). The term
“patient-relevant” relates to the patients` feelings, their functioning and their
survival so usually mortality, morbidity and health-related quality of life (HRQoL)
are considered as patient-relevant outcomes (IQWIG, 2011). In line with § 35b of
the German Social Code Book the following outcomes, which are related to patient
benefit, are especially considered by the IQWIG: Improvement of health status,
reduction of disease duration, increase in life expectancy, reduction of adverse
effects and improvement of quality of life (IQWIG, 2011; SCB V § 35b, 2013). For
topic-related definition of patient-relevant outcomes the IQWIG involves individuals
affected by the disease in question and representatives of patient organizations
(IQWIG, 2011).
In contrast to patient-relevant outcomes surrogate parameters are frequently used
in clinical trials, because they provide results in a shorter period of time and with
less efforts and costs (IQWIG, 2011; Wieczorek et al., 2008). A surrogate marker
can be defined as a substitute for a patient-relevant, clinically meaningful endpoint
and is expected to predict the therapeutic effect of an intervention (Katz, 2004).
Table 9 shows patient-relevant outcomes and surrogate markers often used in
clinical trials with diabetic patients.
20
Table 9: Patient-relevant and surrogate endpoints in diabetes (Wieczorek et al.,
2008, p. 131).
Due to the fact that surrogate markers can be misleading the IQWIG only accepts
surrogate endpoints of clinical trials for their health information products if they
have been validated beforehand. However, there is no standard procedure or
generally accepted method for surrogate endpoint validation (IQWIG, 2011).
In respect to the primary prevention of T2DM in high-risk individuals the IQWIG
argues that the prevention or delay of diabetes diagnosis as such might not be a
patient-relevant outcome but should be considered a surrogate marker. The main
reasons for these considerations are that the diagnosis of T2DM relies on cutpoints for hyperglycemia, which have been derived from data and methodologies
with limitations (see chapter 3.1.5) and additionally that health risks increase below
the cut-points for T2DM and even below the cut-points for IGT.
21
3.3.2.4 Method of information retrieval: systematic literature research, screening
and quality assessment
As mentioned before the basis for health information products are usually
systematic reviews which are identified by the IQWIG in a systematic literature
search. Before conducting the systematic literature research, a project outline
is prepared to inform the IQWIG Department of Information Management about
the background and objectives of the research. It also provides information
about the results of an explorative literature search which is used to identify
search terms and formulate a search strategy for the bibliographic databases
(IQWIG, 2011). “As a quality assurance step, it is tested whether the search
strategy developed in this way identifies known relevant primary publications
(test set) with sufficient certainty.” IQWIG, 2011, p. 95). Furthermore, the
project outline defines the following aspects of the search:
1.
The inclusion criteria with regard to
•
target population, intervention, control group and outcome
•
study design
•
formal characteristics of the publication (e.g. language, time of research,
publication type, quality of review)
2.
The databases included in the search.
The defined criteria are summarized in a PICO scheme (IQWIG, 2011).
The results of the systematic search usually comprise a lot of citations which
are not relevant for the research question. Screening and selection of relevant
publications follows a two-step approach. In the first step two independent
reviewers exclude irrelevant publications by using title and abstract information.
Publications which do not meet the inclusion criteria defined in the project
outline are excluded. For the remaining publications full texts are obtained,
which form the decision basis for inclusion or exclusion. Full texts are again
assessed by two independent reviewers and disagreement is consented by
discussion (IQWIG, 2011).
Identified systematic reviews will then be quality assessed with Oxman and
Guyatt´s validated quality index for systematic reviews in order to minimize the
risk of bias and methodological flaws. “The Institute only uses systematic
reviews on the effects of an intervention for their health information if they fulfill
22
certain minimum requirements, which means that they are only allowed to have
few methodological flaws according to the Oxman and Guyatt Index.” (IQWIG,
2011, p. 81) A description of the validation procedure for the Oxman and
Guyatt Index can be found in Oxman & Guyatt, 1991 and an application form of
the Oxman & Guyatt index used by the IQWIG is available in appendix 1.
Systematic reviews which are rated with an O&G Index ≥ 5 are considered to
be of sufficient methodological quality and can form the basis for the IQWIG
health information. “When more than one systematic review of adequate
methodological quality addresses a particular subject or outcome, a further
quality assessment is carried out.” (IQWIG, 2011, p. 81) Aspects considered
inter alia are the main content of the review in relation to the research question
and the comprehensiveness and actuality of the search (IQWIG, 2011).
23
4
Methods
4.1
Literature Research
As described in chapter 3.3.2.4 the search for evidence for IQWIG health
information products follows a predefined methodology. This methodology was
also used for the master thesis and is described in this section in detail. In
order to precisely define the research question a PICO scheme was
developed. This was done in close collaboration and consultation with Dr.
Martina Ehrlich and Dr. Klaus Koch from the IQWIG Department of Health
Information. The PICO scheme is shown in table 10.
Table 10: PICO scheme for the literature research
Inclusion Criteria
Intervention
Adults (≥ 18 years) without diagnosis of diabetes
mellitus before entering study and with elevated
risk of diabetes mellitus type 2 defined by study
(e.g. Impaired Glucose Tolerance, elevated fasting
blood glucose, obesity)
lifestyle intervention including change in diet and/or
physical activity
Control
General advice/usual care, pharmacological
treatment, no treatment
E3
Outcome
Diagnosis of T2DM, period until diagnosis of T2DM,
overall mortality, cardiovascular morbidity and
mortality, microvascular diseases, quality of life or
clinical parameters (reduction in blood glucose,
blood pressure, BMI)
E4
Study Type
Meta-analysis or systematic review of controlled
studies (CT`s), HTA
E5
Time of Research
The research was conducted in 2009 or later.
E6
Publication
Language
English or German
E7
Publication
Full-text publication available/ procurable
E8
Oxman & Guyatt
Quality of Reviews according to O & G ≥ 5
E9
Scope
Topic / issue relevant
E10
Population
E1
E2
Minimal inclusion criteria: E1-E8 plus E10
As pointed out no generally accepted definition of high-risk individuals for
T2DM exists. Therefore the definition of the population in the PICO allowed for
a broad range of groups “as defined by study”. Also the defined outcomes were
24
quite broad and included several outcomes of interest: For the research question of this master thesis prevention and/or delay of diabetes diagnosis was the
most important outcome. The IQWIG institute was especially interested in patient relevant outcomes like overall mortality, cardiovascular morbidity and mortality, microvascular diseases and quality of life. Clinical parameters (surrogate
endpoints) like reduction in blood glucose, blood pressure and BMI were included, because before the search was carried out it was unclear whether or
not enough high-level evidence on the other outcomes would be retrieved.
The systematic literature research was conducted by the IQWIG Department of
Information Management. To precisely inform the staff about the background and
intention of the research project, a project outline was written (see appendix 2).
This outline included information about the project topic, the definition of lifestyle
changes in the context of this research, the PICO, the databases used for an explorative literature research and the results of this research. The explorative literature research was intended to be used as a test set for the systematic literature
research. The project outline also defined the online databases that were to be
used for the systematic literature review. These were Medline (Ovid), Cochrane
Library, DARE and PubMed.
4.2
Screening, selection of reviews and grading with the Oxman & Guyatt Index
The literature research identified 710 data records. They were independently
reviewed based on title and abstract information by Martina Ehrlich (IQWIG)
and Karin Riemann-Lorenz. Disagreement was found in 42 cases and a consensus was arrived at through discussion. For 47 data records full texts were
obtained and again independently reviewed. One additional systematic review
and meta-analysis was identified through a manual search of the NICE website. As a first step all articles were analyzed for meeting the inclusion criteria
E1-E8 and E10. Nine reviews met these criteria and were quality assessed according to the Oxman and Guyatt Index by Martina Ehrlich and a second reviewer of the IQWIG. Five of the nine systematic reviews yielded an Oxman
and Guyatt Index below 5 and thus have been excluded. In the end 4 systematic reviews of controlled trials that matched all the inclusion criteria defined
in the PICO scheme were identified. A flow chart of the screening process is
shown in Figure 2.
25
Search in Medline (Ovid), Cochrane
Library, DARE, PubMed in June
2013
710 datasets retrieved
Title and abstract analysis by
two independent reviewers
Article identified
through manual
search: 1
Excluded: 663
Full text analysis: 47 + 1
Excluded: 39
E1: 8
E5: 7
E6: 8
E7: 1
E10: 15
Quality assessment according
to Oxman and Guyatt Index: 9
Excluded: 5
Included systematic reviews: 4
Figure 2: Flow chart of screening process
26
A list of the 44 full text articles, which were excluded in the second step of the screening
process, as well as the reasons for exclusion can be found in appendix 3.
5
Results
5.1
Identified Systematic Reviews
The systematic literature research, screening process and quality assessment
identified the following 4 systematic reviews:
•
Dunkley A.J., Charles, K., Gray, L.J., Camosso-Stefinovic, J., Davies, M.J.,
Khunti, K. (2012) Effectiveness of interventions for reducing diabetes and
cardiovascular disease risk in people with metabolic syndrome: systematic
review and mixed treatment comparison meta-analysis. Diabetes, Obesity and
Metabolism, 14(7), 616-625.
•
Jones, R., Freeman, C., Johnson, M., Stevens, J., Buckley Woods, H.,
Guillaume, L., Gillies, C., Goyder, E., Chilcott, J., Payne, N. (2011) Preventing
the progression of pre-diabetes to type 2 diabetes in adults. Systematic review
and meta-analysis of lifestyle, pharmacological and surgical interventions,
ScHARR Public Health Collaborating Centre, retrieved from
http://www.nice.org.uk/guidance/index.jsp?action=download&o=57043 on 20th
of September 2013
•
LeBlanc, E.S., O'Connor, E., Whitlock, E.P., Patnode, C.D., Kapka, T. (2011).
Screening for and Management of Obesity and Overweight in Adults.
Evidence Report No. 89. AHRQ Publication No. 11-05159-EF-1. Rockville,
MD: Agency for Healthcare Research and Quality; October 2011
•
Sumamo, E., Ha, C., Korownyk, C., Vandermeer, B., Dryden, D.M.(2011).
Lifestyle interventions for four conditions: type 2 diabetes, metabolic
syndrome, breast cancer, and prostate cancer. Rockville, MD, USA: Agency
for Healthcare Research and Quality. AHRQ Technology Assessment
Program. 2011.
The main characteristics and results of the systematic reviews are shown in table
11.
27
Table 11: Main characteristics and results of the systematic reviews included
Review
Time of
Search
Specific Research Question
(PICO)
Dunkley et
al., 2012
01/2010
Effectiveness of interventions
for reducing diabetes and
cardiovascular disease risk in
people with metabolic
syndrome.
P: individuals with metabolic
syndrome
I: Lifestyle (Diet and Exercise),
pharmacological therapy or
surgery
C: placebo, usual care or active
control
O: Incidence of T2DM,
Cardiovascular Disease,
reversal of metabolic syndrome
Included Studies with
Lifestyle Intervention
on primary prevention
of T2DM
Meta-analysis
performed for 13 studies
(lifestyle and
pharmacological) on
reversal of metabolic
syndrome.
O&G
Score
Results /Conclusions of
the authors
Ability to answer the
research question/
Limitations
6
Score: Limitations:
• Population restricted to
individuals with metabolic
syndrome.
Only two studies
included, which reported
on T2DM incidence:
LI and pharmacological
interventions can reverse
metabolic syndrome, but it
remains unclear, if these
benefits are sustained
and translate into longer
term prevention of T2DM
and/or CVD. LI appear to
be the most clinically
effective.
• Bo et al., 2007
• Ramachandran et
al., 2007 (IDPP)
Evidence for reduction of
T2DM incidence was
found in two LI trials. .
• Primary outcome
“prevention of T2DM”
only in 2 of 16 studies.
• 3 studies with outcome
“CVD event or mortality”
but limited to
pharmacological
interventions
Minimum follow up of 24 weeks
Abbreviations: LI = Lifestyle interventions, DPP=Diabetes Prevention Program, DPPRG = Diabetes Prevention Program Research Group,
DPS=Diabetes Prevention Study, IDPP= Indian Diabetes Prevention Program
28
Table 11: Main characteristics and results of the systematic reviews included (con`t)
Review
Time of
Search
Specific Research Question
(PICO)
LeBlanc et
al., 2011
AHRQ
09/2010
Effectiveness and harms of
primary care-relevant weightloss interventions for
overweight and obese adults.
P: overweight or obese adults
I: Behavioral or pharmacological
interventions for weight loss
C: usual care - no personalized
intervention that would overlap
with low-intensity intervention
groups
O: Weight loss and maintenance,
improved health outcomes
(morbidity from several diseases,
emotional and physical
functioning and mortality)
Included Studies
with Lifestyle
Intervention on
primary prevention
of T2DM
2 RCTs:
Knowler et al., 2002
(DPP)
Tuomilehto et al.,
2001 (DPS)
O&G
Score
Results /Conclusions of
the authors
Ability to answer the
research question/
Limitations
6
Behaviorally based LI are
safe and effective for
weight loss and
maintenance.
Score: Limitations:
• Population of overweight
and obese adults is a
broader target group than
high-risk individuals for
T2DM. Not all of them
must be considered as
high-risk individuals.
Behaviorally based
interventions which led to
weight loss (4 to 7 kg)
reduced diabetes
incidence by about 30 to
50%.
Data on effects of weightloss on long-term health
outcomes like CVD or
death are insufficient.
• Primary aim of interventions was “weight
loss” and not prevention
of T2DM – so only 2 of 21
included behavioral
interventions reported on
T2DM incidence.
Abbreviations: LI = Lifestyle interventions, DPP=Diabetes Prevention Program, DPPRG = Diabetes Prevention Program Research Group,
DPS=Diabetes Prevention Study, IDPP= Indian Diabetes Prevention Program
29
Table 11: Main characteristics and results of the systematic reviews included (con`t)
Review
Time of
Search
ScHARR
Review;
Jones et al.,
2011 ,
ScHARR
Public Health
Collaborating
Centre
2011
Specific Research Question
(PICO)
Effectiveness of lifestyle,
pharmacological and surgical
interventions for preventing
the progression of prediabetes to type 2 diabetes in
adults.
P: Individuals with IGT or IGF
I: Lifestyle intervention, drugs or
surgical interventions
C: Standard advice, placebo
O: Progression to T2DM
Included Studies with
Lifestyle Intervention
on primary prevention
of T2DM
13 relevant RCTs:
Jarrett et al., 1979
Pan et al., 1997 (Da
Qing)
Wein et al., 1999
Knowler et al., 2002
(DPP)
DPPRG, 2009
Liao, 2002
Lindström et al.,
2003+2006 (DPS)
Kosaka et al., 2005
Ramachandran et al.,
2006 (IDDP)
Roumen et al., 2008
(SLIM)
Penn et al, 2009
Li et al., 2008
Lindahl et al., 2009
O&G
Score
6
Results /Conclusions of
the authors
Each type of lifestyle intervention can reduce the
progress to diabetes in
people with pre-diabetes.
Ability to answer the
research question/
Limitations
Score: +
Limitations: None
In a meta-analysis of 13
studies the pooled HR for
lifestyle interventions was
0,51, 95% CI 0.43-0.62.
A combination of diet and
exercise appears to have
more effect than diet or
exercise alone.
Abbreviations: LI = Lifestyle interventions, DPP=Diabetes Prevention Program, DPPRG = Diabetes Prevention Program Research Group,
DPS=Diabetes Prevention Study, IDPP= Indian Diabetes Prevention Program
30
Table 11: Main characteristics and results of the systematic reviews included (con`t)
Review
Time of
Search
Specific Research Question
(PICO)
Sumamo et
al., 2011
(AHRQ)
03/2010
Effectiveness of lifestyle
interventions to control
progression of T2DM,
progression to DMT2 from
metabolic syndrome or
recurrence of breast and
prostate cancer.
P: Individuals with MS, T2DM,
breast and prostate cancer
patients
I: Exercise and Diet and one
additional component (e.g.
counseling, stress
management).
C: Usual care, diet or exercise
alone or wait list.
O: Progression of existing DMT2
(more medication, CVD etc.)
progression of metabolic
syndrome to DMT2, heart
disease or stroke, recurrence of
breast or prostate cancer.
Included Studies with
Lifestyle Intervention
on primary prevention
of T2DM
4 relevant RCTs that
reported on T2DM
incidence, two of which
also reported on CVD
events:
- Bo et al., 2007
- Knowler et al., 2002
(DPP)
- Eriksson et al., 1999
(DPS)
- Pan et al., 1997, (Da
Qing)
Meta-analyses were
performed for surrogate
parameters like blood
pressure, change in diet
and physical activity,
weight loss etc.
O&G
Score
7
Results /Conclusions of
the authors
Ability to answer the
research question/
Limitations
Development of T2DM was
significantly de-creased in
the LI groups in 4 RCTs. In
two studies this was also
shown in long-term (10 and
20 years).
Score: Limitations:
• Population restricted
to individuals with
metabolic syndrome.
• Considerable heterogeneity of LI,
because additional
component besides
diet and exercise
differed between
interventions.
No significant differences
could be found for CVDevents and mortality in 2
studies (DPS und Da Qing)
For different surrogate
markers advantages could
be shown for the LI groups
(details see review p. 53-54)
Abbreviations: LI = Lifestyle interventions, DPP=Diabetes Prevention Program, DPPRG = Diabetes Prevention Program Research Group, DPS=Diabetes
Prevention Study, IDPP= Indian Diabetes Prevention Program
31
5.1.1 Overall results of identified systematic reviews
In general all systematic reviews came to the conclusion that lifestyle interventions can reduce diabetes incidence in high-risk individuals. LeBlanc et al.
stated that behaviorally based interventions, which led to weight loss of 4 to 7
kg, reduced diabetes incidence by about 30 to 50% (LeBlanc et al., 2011).
Sumamo et al. reported that development of T2DM was significantly decreased
in the LI groups in 4 RCTs. In two studies this held true in long-term follow-up
(10 and 20 years) (Sumamo et al., 2011). The systematic review of Dunkley et
al. concluded that LI and pharmacological interventions can reverse metabolic
syndrome, but that it remained unclear if these benefits are sustained and
translate into longer term prevention of T2DM and/or CVD. LI appeared to be
the most clinically effective. Evidence for reduction of T2DM incidence was
found in two LI trials (Dunkley et al., 2012). The results of the ScHARR review
will be described in detail in chapter 5.1.2.
However, the overview table also reveals that the systematic reviews showed
considerable differences, most of which can be attributed to the broad definitions of population and outcomes in the PICO used for the systematic literature
research. Two of the systematic reviews (Dunkley et al., 2012 and Sumamo et
al., 2011) used the metabolic syndrome as the qualifying condition to identify
individuals at high risk of developing T2DM. The systematic review of LeBlanc
et al. included RCTs that considered overweight and obese individuals as
having a high risk for diabetes whereas the ScHARR review included all RCTs
that defined high-risk individuals by IGT (LeBlanc et al., 2011, Jones et al.,
2011). Furthermore, it becomes obvious that in addition the range of included
outcomes differs among the reviews. Besides incidence of T2DM the
systematic reviews of Dunkley et al., 2012, LeBlanc et al., 2011 and Sumamo
et al., 2011 also listed surrogate parameters like weight loss and physical
functioning, reversal of metabolic syndrome and endpoints like mortality, CVD
mortality and CVD morbidity as outcomes of interest in their PICO schemes.
Three systematic reviews (Dunkley et al., 2012; the ScHARR Review 2011 and
LeBlanc et al., 2011) assessed not only the effectiveness of LI but also the
effectiveness of pharmacological interventions for different outcomes. As the
authors provided information on LI separately from information about drug
interventions, it was nevertheless possible to include the reviews.
32
Consequently the systematic reviews also differ in the RCTs included in their
investigations. Dunkley et al. included only two studies, which reported on
T2DM incidence (Bo et al., 2007 and Ramachandran et al., 2007 (IDPP)).
Sumamo et al. (AHRQ) found 4 relevant RCTs that reported on T2DM incidence (Bo et al., 2007; Knowler et al., 2002 (DPP); Eriksson et al., 1999 (DPS);
Pan et al., 1997, (Da Qing)), two of which also reported on CVD events.
LeBlanc et al. included 2 RCTs (Knowler et al., 2002 (DPP) and Tuomilehto et
al., 2001 (DPS)). In contrast the ScHARR review by Jones et al. discovered 13
RCTs that matched their inclusion criteria (details see table 11).
In summary one can say that due to the broad definitions of population and
outcomes used in the PICO for the systematic literature research of this master
thesis, the retrieved systematic reviews differ in the specific research question
they try to answer and in scope.
In order to identify the systematic review which has the best ability to answer
the research question of this thesis, scoring has been necessary. Systematic
reviews, which fell within the PICO but had limitations in answering the
research question, were scored: - . Limitations could inter alia be
•
a narrow definition of the study population, which resulted in evaluating the
effectiveness of LI on the prevention of T2DM in only a certain subgroup of
high-risk individuals or a very broad definition which led to the inclusion of
subjects other than high-risk individuals in the RCTs
and/or
•
prevention of T2DM being not the only and not the most important primary
outcome of the systematic review.
Systematic reviews that had no such limitations were scored: + and considered
the most suitable for answering the research question of this master thesis.
The result of this assessment is provided in table 11, last column. The
assessment shows that the ScHARR review performed by Jones et al. in 2011
is considered to be the most suitable to answer the primary research question
of this master thesis. Hereafter only the results of this systematic review will be
provided in detail.
33
5.1.2 The ScHARR Review
The ScHARR Review by Jones et al. comprises several aspects, which are not
within the scope of this master thesis. Among them are a meta-analysis of the
effectiveness of pharmacological interventions, an assessment of the effectiveness of interventions especially for South Asian populations and an estimation
of a treatment ranking, which might be interesting for the evidence based
patient information by the IQWIG. However, these aspects will not be
discussed in this master thesis.
5.1.2.1 Methods
Jones et al. performed a systematic literature research in several databases
(Medline In Process and Other Non Indexed Citations and Medline 1950-Current
via OVID SP, Embase via OVID SP, Cochrane Library (DARE, CENTRAL, HTA)
via Wiley, CINAHL via EBSCO, BNI via OVID, Science and Social Science Citation
Indices via Web of Knowledge, PsycINFO via OVID SP and EPPI Centre) and
additionally searched for grey literature and on websites of relevant organizations
(e.g. National Library for Public Health, Diabetes and Obesity Research Network
(DORN)) (Jones et al., 2011). To be considered in the review studies needed to
include subjects with pre-diabetes, lifestyle, drug or surgical interventions and
development of T2DM as a required outcome measure. Search was limited to
RCTs. Screening was done by two independent reviewers. In the end 26
articles were included in the review. All articles were quality assessed with the
Jadad score and additionally with the NICE quality assessment criteria checklist. On the whole the authors judged the quality of papers as very good, with
21 papers being rated as very good (++), three as good (+) and two as poor (-).
Thirteen studies or study subgroups from 11 RCTs, which provided enough data,
were included in a meta-analysis.
5.1.2.2 Included RCTs
Basic characteristics and results of the RCTs included in the meta-analysis can be
found in table 12. It becomes obvious that there is considerable clinical heterogeneity between the different RCTs in terms of study population, intervention characteristics, follow-up period and quality of the studies, which will be briefly summarized in the following section.
34
Table 12: Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al., 1979,
Knowler et al., 2002
Study (RCT)
Population Characteristics
Intervention
Study/Author: Da Qing Study/ Pan et al., 1997
Country: China
Quality: NICE ++
No of participants: 577
Definition of high risk: IGT
Ethnicity: Chinese
Mean age: 45 +/- 9.1
Mean BMI (kg/m2): 25.8 +/- 3.8
% BMI ≥ 30: NR
Intervention groups: diet only, exercise only, diet & exercise
Lifestyle change:
Diet: Low fat, high carbohydrate diet – weight loss recommended for people with BMI > 25 at a rate of 0.51.0 kg per month until they achieved a BMI of 23; participants were encouraged to consume more
vegetables, control their intake of alcohol, and reduce their intake of simple sugars.
Exercise: participants were encouraged to increase the amount of leisure physical exercise by at least one
unit per day (such as slow walking for 30 minutes, fast walking for 20 minutes etc) and by two units per day
if possible for those <50 years of age with no evidence of cardiovascular disease or arthritis.
Control
Randomization
Follow- up
Study attrition
Administration of advice: individual and small group counseling; sessions were conducted weekly for one
month, monthly for three months, and then once every three months for the remainder for 6 years
Behavioral change strategies: NR
General information about lifestyle and diabetes risk. No individual or group counseling.
by clinical centre; cluster-randomization
6 years
8.1%
35
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Primary Outcomes
Study/Author: Da Qing Study/ Pan et al., 1997 con`t
Country: China
Quality: NICE ++
Incidence of T2DM:
Diet: 10.0 (95% CI, 7.5-12.5) per 100 person years
Exercise: 8.3 (6.4-10.3) per 100 person years
Diet & Exercise: 9.6 (7.2-12.0) per 100 person-years
Control: 15.7 (95% CI 12.7-18.7) per 100 person-years
Relative Risk Reduction compared to Control after 6 years:
Diet: – 33%
Exercise: – 47%
Diet & Exercise: - 38%
Cumulative incidence of T2DM after 6 years
Exercise: 41%
Diet 44%
Diet & Exercise: 46%
Control: 68%
Weight loss: Only the diet & exercise group had a mean weight loss of 1.77 kg , the other groups gained
Secondary Outcomes
weight
Authors Conclusions
• Diet and/or exercise interventions led to a significant decrease in the incidence of diabetes over a six
year period among those with IGT.
• No benefit in terms of CVD or all-cause mortality maybe do to a lack of power
PA= Physical Activity; NR = Not reported; CH= carbohydrate
36
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Population
Characteristics
Intervention
Control
Randomization
Average follow- up
Study attrition
Primary Outcomes
Study/Author: DPP/ Knowler et al., 2002
Country: US
Quality: NICE ++
No of participants: 3234
Definition of high risk: IGT
Ethnicity: 54,7% white, 19.9 % African, 15.7 % Hispanic, 5.3 % Am. Indian, 4.4% Asian
Mean age: 50.6 +/- 10.7 y
Mean BMI (kg/m2): 34 kg/m2
% BMI ≥ 30: 67,7%
Intervention groups: diet & exercise AND standard advice + metformin
Lifestyle change: Low fat, low calorie diet and moderate PA (≥ 150 min./week) and 7% weight loss
Administration of advice: 16 individual sessions in 24 weeks, during maintenance phase offer of group
sessions
Behavioral change strategies: Yes. Goal-setting, self-monitoring, stimulus control, problem-solving and
relapse prevention.
Standard advice + placebo
by clinical centre; cluster-randomization
2.8 years
8%
Incidence of T2DM:
Diet & Exercise: 4.8 per 100 person years
Metformin: 7.8 per 100 person years
Control: 11 per 100 person years
Relative Risk Reduction compared to Control:
Diet & Exercise: - 58% (95% CI 48%-66%)
Metformin: -31% (95% CI 17%-43%)
Cumulative incidence of T2DM after 3 years:
Diet & Exercise: 14.4%
Metformin: 21.7%
Placebo: 28.9%
NNT after 3 years: Diet & Exercise: 6.9 (95% CI 5.4-9.5); Metformin: 13.9 (95% CI 8.7-33.9)
37
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Secondary Outcomes
Authors Conclusions
Study/Author: DPP/ Knowler et al., 2002 con`t
Country: US
Quality: NICE ++
• The advantage of LI over Metformin was greater in older persons and those with a lower BMI than in
younger persons and those with a higher BMI.
• 50% of the participants in the LI group achieved the goal of a 7% or more weight loss by the end of 24
weeks.
• Participants in the lifestyle intervention group had a greater weight loss and a greater increase in leisure
activity than the participants in the metformin and placebo groups.
• The average weight loss was 0.12 kg in the placebo group, 2.1 kg in the metformin group and 5.6 kg in the
lifestyle intervention group
Lifestyle changes and treatment with Metformin both reduced the incidence of diabetes in persons at high risk.
The lifestyle intervention was more effective than Metformin.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
38
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Study/Author: Diabetes Prevention Program Outcomes Study (DPPOS)/ Diabetes Prevention Program Research
Group, 2009
Country: US
Quality: NICE ++
Population
No of participants: 2766
Characteristics
Definition of high risk: follow up of eligible participants from DPP with IGT
Ethnicity: Mean age: 55.2 +/- 10.3 y
Mean BMI (kg/m2): male: 31.1 + 5.9; female: 34.2 + 7.2
% BMI ≥ 30: Intervention groups: diet & exercise AND standard advice + Metformin
Intervention
Lifestyle change: Low fat, low calorie diet and moderate PA (≥ 150 min./week)and 7% weight loss
Administration of advice: 16 individual sessions in 24 weeks, during maintenance phase offer of group sessions
Behavioral change strategies: Yes. Goal-setting, self-monitoring, stimulus control, problem-solving and relapse
prevention.
Control
Standard advice + placebo
Randomization
by clinical centre; cluster-randomization
Average follow- up
10 years after DPP randomization
Study attrition
Relative Risk Reduction compared to Placebo:
Primary Outcomes
Diet & Exercise: - 34% (95% CI 24%-42%); Metformin: -18% (95% CI 7%-28%)
The lifestyle effect was greatest in participants aged 60–85 years at randomization (49% risk reduction), in whom
metformin had no significant effect. The median delay to onset of diabetes was approximately 4 years by lifestyle
and 2 years by metformin, compared with placebo.
Secondary Outcomes At the most recent yearly examination, 23% in the lifestyle, 19% in the metformin, and 19% of participants in the
placebo groups had become normoglycaemic by criteria defined and reported previously (fasting glucose <6.1
mmol/L, 2-h glucose <7.8 mmol/L, and no previous diagnosis of diabetes.
Authors Conclusions
During follow-up after DPP, incidences in the former placebo and metformin groups fell to equal those in the
former lifestyle group, but the cumulative incidence of diabetes remained lowest in the lifestyle group. Prevention
or delay of diabetes with lifestyle intervention or metformin can persist for at least 10 years.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
39
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Population Characteristics
Intervention
Control
Randomization
Average follow- up
Study attrition
Primary Outcomes
Study/Author: DPS/ Lindström et al., 2003 and update 2006
Country: Finland
Quality: NICE ++
No of participants: 522
Definition of high risk: IGT
Ethnicity: Mean age: 55 +/- 7 years
Mean BMI (kg/m2): 31
% BMI ≥ 30: 96. 5% ♂; 66.3% ♀
Intervention groups: Diet & Exercise
Lifestyle change: Low-fat, high fibre diet, increase of PA and ≥ 5% weight loss
Administration of advice: 15 individual sessions over 3 years with nutritionist. Group activities like cooking
and supermarket visits encouraged. Printed material was used to illustrate the messages and to serve as a
reminder at home. Individualized and supervised exercise sessions. Endurance exercise and moderate
intensity resistance training sessions were also offered free of charge.
Behavioral change strategies: Yes. Problem-solving
General information about lifestyle and diabetes risk, but not individualized.
Individual randomization at the first study visit after the screening phase
3.2 years (active phase and post-intervention follow-up: 7 years)
8%
Incidence of T2DM after mean follow up of 7 years:
Diet & Exercise: 4.3 per 100 person years
Control: 7.4 per 100 person years
Relative Risk Reduction after mean post follow up of 7 years:
Diet & Exercise: - 42%
Cumulative incidence of T2DM after 6 years:
Diet & Exercise: 23%
Control: 38%
Absolute risk reduction of 15% (CI 7.2–23.2)
NNT: 22 for 1 year
40
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Secondary Outcomes
Authors Conclusions
Study/Author: DPS/ Lindström et al., 2003 and update 2006 con`t
Country: Finland
Quality: NICE ++
• weight loss was greater in the IG than in the CG after 3 years: IG: -3.5 kg ; CG: .- 0.9 kg
• 10 % in the intervention group and 27% in the control group did not achieve any of the predefined goals
by the 3-year examination
• 14% in the intervention and 6% in the control group achieved four or five goals.
• There was a strong inverse correlation between the success score and the incidence of diabetes during
the total follow-up.
Lifestyle intervention in people at high risk for type 2 diabetes resulted in sustained lifestyle changes and a
reduction in diabetes incidence, which remained after the individual lifestyle counseling was stopped.
No benefit in terms of CVD or all-cause mortality maybe do to a lack of power
PA= Physical Activity; NR = Not reported; CH= carbohydrate
41
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Study/Author: IDDP/ Ramachandran et al., 2006
Country: India
Quality: NICE ++
Population Characteristics No of participants: 531
Definition of high risk: IGT
Ethnicity: Asian Indians, middle class
Age range: 35-55 years
Mean BMI (kg/m2): 25.6 – 26.3 +/- 3.7
% BMI ≥ 30: Intervention groups: LI; LI + Metformin; Metformin only
Intervention
Lifestyle change: Reduction of total calorie, refined CH and fats + increase of fibre. Individual advice on PA.
Aim: ≥ 30 min. of brisk walking per day
Administration of advice: Monthly telephone calls and individual sessions every 6 months for lifestyle
advice
Behavioral change strategies: NR
Control
Standard advice
Randomization
Average Follow- up
3 years
Study attrition
5%
Relative Risk Reduction after 3 years:
Primary Outcomes
Diet & Exercise: - 28.5%
Diet & Exercise + Metformin: -26.4 %
Metformin only:- 28.2 %
Cumulative incidence of T2DM after 3 years
Diet & Exercise: 39.3%; Diet & Exercise + Metformin: 39.5 %
Metformin only: 40.5%; Control: 55 %
NNT after 3 years:
Diet & Exercise: 6.4 ; Diet & Exercise + Metformin: 6.9; Metformin only: 6.5
Secondary Outcomes
No significant weight loss in the IG`s – reduction of risk occurred without weight loss
Authors Conclusions
LI showed significant reduction of diabetes incidence after 3 years from 55% to 40 %.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
42
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Population Characteristics
Intervention
Control
Randomization
Average follow- up
Study attrition
Primary Outcomes
Study/Author: Jarrett Study/ Jarrett et al. 1979
Country: UK
Quality: NICE +
No of participants: 204 (all male)
Definition of high risk: Survey blood sugar elevated and IGT (for details see Jarrett et al., 1979)
Ethnicity:Mean age: Mean BMI (kg/m2): % BMI ≥ 30: Intervention groups: 1) low CH diet (120g CH/d) + placebo; 2) “limit sucrose intake” + placebo 3) low CH
diet (120g CH/d) + 50 mg phenformin S.A. /d 4) “limit sucrose intake” + 50 mg phenformin S.A. /d
Lifestyle change: The LI groups + placebo served as control groups
Administration of advice: The diet was taught with a specially developed booklet. Intensity is NR.
Behavioral change strategies: NR
1) low CH diet (120g CH/d) + placebo AND 2) “limit sucrose intake” + placebo
Individually at first visit to clinic
5 years
“Worsening to diabetes” which was arbitrarily defined by the study authors (for details see Jarrett et al.,
1979):
1) low CH diet (120g CH/d) + placebo : 18.2% (8/44)
2) “limit sucrose intake” + placebo: 13.3% (6/45)
3) low CH diet (120g CH/d) + 50 mg phenformin S.A. /d: 9.3% (4/43)
4) “limit sucrose intake” + 50 mg phenformin S.A. /d: 18.4% (9/49)
Secondary Outcomes
Authors Conclusions
The outcome „worsened to diabetes“ was not significantly different in any of the treatment groups.
Treatment was not predictive for the outcome.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
43
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Study/Author: Kosaka Study/ Kosaka et al., 2005
Country: Japan
Quality: NICE ++
Population Characteristics
No of participants: 485 (all male, 356 control, 102 intervention)
Definition of high risk: IFG + IGT (1980 WHO criteria for prediabetes)
Ethnicity: Japanese
Mean age: NR, age range: 87% between 40 and 60 y
Mean BMI (kg/m2): 24 kg/m2
% BMI ≥ 30:Intervention groups: diet & exercise
Intervention
Lifestyle change: Individually tailored dietary advice (fat, alcohol, calorie intake, vegetables) and
recommendation of moderate intensity leisure time PA. Weight reduction was recommended to a BMI of 22
Administration of advice: individual advice every 2-3 months
Behavioral change strategies: NR
Control
Standard advice on lifestyle change every 6 months
Randomization
Individuals randomly assigned to IG or CG
Follow- up
4 years
Study attrition
4.7 – 5.5 %
Cumulative incidence of T2DM after 4 years
Primary Outcomes
Diet & Exercise: 3% (3/102)
Control: 9.3% (33/356)
Regression to NGT:
IG: 53 %; CG: 33.9%
Weight loss:
Secondary Outcomes
Diet & Exercise: - 2.18 kg
Control: -0.39 kg
Authors Conclusions
A lifestyle intervention designed to achieve and maintain ideal body weight (BMI <22 kg/m2) is an effective
means of reducing incidence of type 2 diabetes in Japanese males with IGT.
Reduction in the incidence by lifestyle intervention was successfully carried out in a clinical outpatient
setting for diabetic patients.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
44
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Study/Author: The China Da Qing Diabetes Prevention Study/ Li et al., 2008 (follow-up from Da Qing Study)
Country: China
Quality: NICE Population Characteristics
No of participants: 577
Definition of high risk: IGT
Ethnicity: Chinese
Mean age: 45 +/- 9.1
Mean BMI (kg/m2): 25.8 +/- 3.8; % BMI ≥ 30: NR
Intervention groups: diet only, exercise only, diet + exercise
Intervention
Lifestyle change:
Diet: Low fat, high carbohydrate diet – weight loss recommended for people with BMI > 25 at a rate of 0.51.0 kg per month until they achieved a BMI of 23; participants were encouraged to consume more
vegetables, control their intake of alcohol, and reduce their intake of simple sugars.
Exercise: participants were encouraged to increase the amount of leisure physical exercise by at least one
unit per day (such as slow walking for 30 minutes, fast walking for 20 minutes etc) and by two units per day
if possible for those <50 years of age with no evidence of cardiovascular disease or arthritis.
Administration of advice: individual and small group counseling; sessions were conducted weekly for one
month, monthly for three months, and then once every three months for the remainder for 6 years
Behavioral change strategies: NR
Control
General information about lifestyle and diabetes risk. No individual or group counseling.
Randomization
By clinical centre /cluster-randomization
Follow- up
20 years
Study attrition
Cumulative incidence of T2DM after 20 years:
Primary Outcomes
IG: 80%; CG: 93%
NNT: 6
Secondary Outcomes
Participants in the intervention group had an average of 3.6 fewer years with diabetes
Authors Conclusions
Group-based lifestyle interventions over 6 years can prevent or delay diabetes for up to 14 years after the
active intervention. However, whether lifestyle intervention also leads to reduced CVD and mortality remains
unclear.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
45
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Population Characteristics
Intervention
Control
Randomization
Average follow- up
Study attrition
Primary Outcomes
Study/Author: Liao Study/ Liao et al., 2002
Country: US
Quality: NICE +
No of participants: 64
Definition of high risk: IGT
Ethnicity: Japanese
Mean age: IG: 55.8 years ± 1.8; CG: 52.2 years ± 1.8
Mean BMI (kg/m2): IG: 25.6 ± 0.8; CG: 26.6 ± 0.8
% BMI ≥ 30: Intervention group: diet & exercise
Lifestyle change: Endurance exercise training and a low fat, high CH diet (<30% of total calories as fat
(<7% as saturated fat), 55% as carbohydrate, the balance as protein, and <200 mg cholesterol daily.)
Administration of advice: exercise training was directed by an exercise physiologist for the first 6
months.
Behavioral change strategies: NR
Supervised stretching exercise training and a “healthy diet” (30% of total calories as fat (10% as
saturated fat), 50% as carbohydrate, 20% as protein, and <300 mg cholesterol daily.)
Adaptive randomization
24 months
Study was not designed to demonstrate prevention of diabetes
Cumulative incidence of T2DM after 6 months:
one person in each group had developed diabetes.
Cumulative incidence T2DM after 12 months:
one person from the intervention group had diabetes (1/32, 3.1%), and two from the control group had
diabetes (2/32, 6.3%)
Secondary Outcomes
Weight loss:
At 6 and 12 months the IG showed greater weight loss than the CG.
Authors Conclusions
Diet and endurance exercise improved BMI, body composition, and body fat distribution and, thus, may
delay or prevent type 2 diabetes in Japanese Americans with IGT.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
46
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Population Characteristics
Intervention
Control
Randomization
Follow- up
Study attrition
Study/Author: Lindahl Study/ Lindahl et al., 2009
Country: Sweden
Quality: NICE ++
No of participants: 301
Definition of high risk: IGT and BMI > 27 kg/ m2
Ethnicity: Mean age: IG: 52.2 +9.0 y, CG: 53.5 + 8.4 y
Mean BMI (kg/m2): IG: 31.2 + 3.1, CG: 30.2 +3.4
% BMI ≥ 30: Intervention groups: diet & exercise
Lifestyle change: 1 month residential intensive program with ~ 140 h of scheduled activities. Moderate
intensity PA training 2.5 h/d and low-fat and high fibre diet. No alcohol allowed and smoking cessation
program offered. Additional learning sessions were repeated during a 4-day follow-up 12 months later.
Administration of advice:Behavioral change strategies: Yes, goalsetting, self-monitoring, problem-solving, stress management
and relapse prevention
A health survey was performed, including a physical examination, a 2-h OGTT and blood sampling. The
survey was followed by a 30–60-min counseling session, where the participants were given both oral
and written advice. The same examination protocol was repeated after 1, 3 and 5 years.
Individual randomization to IG or CG
1, 3 and 5 years
47
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Primary Outcomes
Secondary Outcomes
Authors Conclusions
Study/Author: Lindahl Study/ Lindahl et al., 2009 con`t
Country: Sweden
Quality: NICE ++
Incidence of T2DM at 1 year:
LI: 7%; CG: 25%; RRR: 70%
Incidence of T2DM at 3 years:
LI: 17%; CG: 25%; RRR: 40%
Incidence of T2DM at 5 years:
LI: 24%; CG: 29%; RRR: 25%
The risk reductions at 3 and 5 years were not significant.
The intervention affected several important cardio-metabolic risk variables beneficially and reduced the
risk for type 2 diabetes, but the effects persisted only as long as the new lifestyle was maintained.
Increased physical activity seemed to be the behaviour that was most easy to preserve.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
48
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Population Characteristics
Intervention
Control
Randomization
Mean follow-up
Study attrition
Primary Outcomes
Secondary Outcomes
Authors Conclusions
Study/Author: Penn Study/ Penn et al., 2009
Country: UK
Quality: NICE ++
No of participants: 102
Definition of high risk: IGT
Ethnicity:
Mean age: IG: 56.8 y, CG: 57.4 y
Mean BMI (kg/m2): IG: 34.1, CG: 33.5
% BMI ≥ 30: Intervention group: diet & exercise
Lifestyle change: high CH and fibre, low fat diet to achieve weight loss to a BMI < 25 kg/m2. Moderate
PA of 30 minutes/d was encouraged.
Administration of advice: individual advice and motivational interviewing
Behavioral change strategies: NR
Minimal intervention control group
Individual randomization stratified by sex and by 2 hour plasma glucose value
3.11 years
Incidence of T2DM:
IG: 32.7 (95% CI 10.7 to 74.6) per 1000 person years of follow-up
CG: 67.1 (95% CI 34.2 to 117.5) per 1000 person years of follow-up
RR: 0.45 (95% CI 0.2 to 1.2)
Weight loss:at year 1 follow-up IG had a significantly higher mean weight change: -2.3 kg than CG: 0.01 kg. Only three participants achieved BMI < 25 kg/m2.
The results are consistent with other diabetes prevention trials. This study was designed as part of a
larger study and although the sample size limits statistical significance, the results contribute to the
evidence that T2DM can be prevented by lifestyle changes in adults with IGT. In explanatory analysis
small sustained beneficial changes in weight, physical activity or dietary factors were associated with
reduction in T2DM incidence.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
49
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Population Characteristics
Intervention
Control
Randomization
Follow- up
Study attrition
Primary Outcomes
Study/Author: SLIM Study/ Roumen et al., 2008
Country: The Netherlands
Quality: NICE ++
No of participants: 147
Definition of high risk: IGT
Ethnicity:Mean age: IG: 54.2 + 5.8 y; CG: 58.4 + 6.8 y
Mean BMI (kg/m2): IG: 29.6 + 3.8, CG: 29.2 + 3.3
% BMI ≥ 30:Intervention group: diet & exercise
Lifestyle change: Dietary recommendations based on the Dutch guidelines for a healthy diet and increase
level of PA to at least 30 min. a day for at least 5 days a week. Weight loss of 5–7% was another objective.
Administration of advice: Individual advice on how to reach dietary and PA goals. Participation in a
combined aerobic and resistance exercise program was encouraged.
Behavioral change strategies: self-monitoring , goal-setting
Standard advice on the benefits of lifestyle change, but no individual counseling.
Individual randomization with stratification for sex and mean 2-h plasma glucose concentration.
Data analyses of the 3-year results include those subjects still participating in the study (completers, n= 106:
52 IG subjects and 54 CG subjects).
28% (41 dropouts of 147 subjects origionally included)
Cumulative incidence of T2DM after 3 years (completers):
IG: 18% (8/44); CG: 38% (18/47)
The P-value of the log-rank test was 0.025, and the relative risk was 0.42 [95% confidence interval (CI) 0.18–0.96]
Cumulative incidence in the intention to treat analysis:
IG: 18% (11/61); CG: 32% (19/60)
The P-value from the log-rank test was 0.07 and the relative risk 0.52 (95% CI 0.25–1.10).
50
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al., 1979,
Knowler et al., 2002 (con`t)
Study (RCT)
Study/Author: SLIM Study/ Roumen et al., 2008 (con`t)
Country: The Netherlands
Quality: NICE ++
Secondary Outcomes
Decrease in body weight correlated with a decrease in 2 h glucose levels.
Authors Conclusions
The lifestyle intervention showed a sustained beneficial effect on 2-h glucose concentrations, insulin
resistance and 2-h FFA, even after 3 years. The lifestyle intervention is effective, but for implementation more
information is needed about factors influencing adherence.
Comments
The authors noted that incidence of diabetes was not the primary outcome of the study and although it was
examined, the results have to be interpreted with caution because the study was underpowered for those
analyses.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
51
Table 12:
Basic characteristics of included RCTs – information compiled from Gillett et al. 2012, and Jones et al., 2011; Jarrett et al.,
1979, Knowler et al., 2002 (con`t)
Study (RCT)
Population Characteristics
Intervention
Control
Randomization
Mean follow- up
Study attrition
Primary Outcomes
Secondary Outcomes
Authors Conclusions
Study/Author: Wein Study/ Wein et al., 1999
Country: Australia
Quality: NICE No of participants: 200 (all female)
Definition of high risk: IGT
Ethnicity:
Mean age: IG: 39.5 years (95% CI 38.2 to 40.8) ; CG: 37.8 years (95% CI 36.5 to 39.0)
Mean BMI (kg/m2): IG: 25.2 (95% CI 24.1 to 26.4); CG: 25.6 (95% CI 24.5 to 26.8)
% BMI ≥ 30:Intervention group: diet
Lifestyle change: group received dietary questionnaires and standard diet advice sheet. Telephone contact
with the dietician was arranged three-monthly. The importance of regular exercise (e.g. brisk walking for 30
minutes three times per week) was stated.
Administration of advice:Behavioral change strategies: Like intervention group but without telephone calls.
NR
51 months, IG: 58.6 months, CG: 47.9 months
Cumulative Incidence of T2DM:
IG: 26.8%
CG: 28.1 %
No significant difference
-
In women with impaired glucose tolerance, this randomized controlled study showed no significant benefit in
women given dietary guidelines reinforced with continued, regular contact with a dietician, compared with
those given dietary guidelines alone. However, compared with the Da Qing study, the results in the control
group were encouraging, and combined with the results of other trials, suggest that dietary intervention is
warranted in individuals with impaired glucose tolerance.
PA= Physical Activity; NR = Not reported; CH= carbohydrate
52
Legend:
Study quality (using NICE methodology) (Jones et al., 2011, p. 32)
Grade
++
+
–
Criteria
All or most of the criteria have been fulfilled. Where they have not been fulfilled the conclusions are thought very unlikely to alter.
Some of the criteria have been fulfilled. Those criteria that have not been fulfilled or adequately described are thought unlikely to
alter the conclusions.
Few or no criteria have been fulfilled. The conclusions of the study are thought likely or very likely to alter.
53
Study populations: Studies have been carried out in several different countries
and included populations with varying ethnicity. 5 RCTs included individuals of
Asian ethnicity only and two US studies included populations of mixed ethnicity
(Whites, African, Hispanic, Asian). For the five European studies implemented in
the Netherlands, Finland, Sweden and UK and the only Australian study, details
about ethnicity of populations has not always been reported. The number of
participants in the different RCTs ranged from 64 in a small US study with
individuals of Japanese ancestry only to 3234 participants in the US DPP. Mean
BMI at baseline differed substantially between the different study populations
(range: 24 kg/m2 in the Japanese Kosaka study to 34 kg/m2 in the US DPP).
Generally mean BMI was lower in the study populations with Asian ethnicity than in
the American and European study groups.
Intervention characteristics: Three of the studies reported on a diet only
intervention, one study on an exercise only intervention and nine on a combined
diet and exercise intervention. Change in diet usually comprised a low fat, high
carbohydrate diet with restricted calorie intake in order to achieve weight loss.
However, other diet components like alcohol or sugar restrictions as well as the
targets set for weight loss differed substantially. The same was true for the
exercise component which ranged from merely encouraging participants to have
more leisure time activity to instructed physical activity training of 2.5 h/d in a 1
month residential intensive program.
Also the administration of dietary and physical activity advice differed widely in
terms of duration and intensity between the interventions. In one study (Wein et al.,
1999) the intervention goals were reinforced by three-monthly telephone calls with
a dietician only whereas other participants received individual and small-group
counseling for up to 6 years (Da Qing Study, Pan et al., 1997). The use of
behavioral change strategies like goal-setting, self-monitoring, problem solving and
relapse prevention was reported by 5 studies.
Follow-up-period: Mean follow up of study participants was lowest in the study of
Liao et al. with 2 years (Liao et al., 2002). Three studies included in the metaanalysis provided long-term follow up data of participants. These were Lindström
et al. with a 7 year follow-up of the Finnish DPS (Lindström et al. 2006), the
Diabetes Prevention Program Outcomes Study (DPPOS) with a 10 year follow-up
54
of the US DPP (DPPRG, 2009) and Li et al. with a 20 year follow-up of the Da
Qing study (Li et al., 2008).
Quality of RCTs: As mentioned in the methods part quality assessment was done
with the NICE instrument. Jones et al. classified the included RCTs as follows: 2
studies with NICE: - (poor quality), 2 studies with NICE: + (good quality) and 9
studies with NICE: ++ (very good quality) (Jones et al., 2011). “Study quality did
not determine inclusion into or exclusion from the review.” (Jones et al., 2011, p.
30). The results of the quality assessment for the individual RCTs can be seen in
table 12.
5.1.2.3 Meta-analysis on prevention of T2DM
Jones et al. found that in all studies the LI groups had lower progression rates to
T2DM than the control groups (Jones et al., 2011). The results of the metaanalysis can be seen in Figure 3:
Figure 3: Meta-analysis of lifestyle interventions (Jones et al., 2011, p. 51)
55
As an outcome effect measure Jones et al. calculated HRs with 95% CI; the metaanalysis was performed using a random-effects model. When comparing 13
studies or study arms in a meta-analysis a pooled HR of 0.51 (95% CI 0.43 – 0.62)
was obtained (p-value of 0.00001, using standard meta-analysis with Z-scores).
Figure 3 shows that the HRs of all included studies or study subgroups are left to
the line of no effect which indicates that all studies favor treatment. Confidence
intervals overlap the line of no effect in 4 studies (Jarrett et al., 1979, Wein et al.,
1999, Liao et al., 2002, Penn et al., 2009) which means that in these studies the
effects were not statistically significant. The test for statistical heterogeneity
resulted in an I2 of 42.3 % which indicates moderate heterogeneity among studies
and is in conformity with the differences in basic study characteristics described in
chapter 5.1.2.2.
Figure 3 shows the HR for the different types of interventions (diet only, exercise
only, diet and exercise). Based on three studies (Jarrett et al., 1979, Da Qing
Study (Pan et al., 1997) and Wein et al., 1999) for diet only interventions the HR
was 0.67 (95% CI 0.49-0.92). Only one study arm of the Da Qing Study reported
on an exercise only intervention with a HR of 0.53 (95% CI 0.34 -0.83). Nine
studies provided data for combined diet and exercise interventions with a pooled
HR of 0.47 (95% CI 0.37 – 0.59). The pooled HR of all LI was 0.51 (95% CI 0.43 –
0.62) and statistically significant (Jones et al., 2011).
The authors concluded that “…lifestyle interventions have an effect in delaying or
preventing progress to diabetes in people with IGT.” and “…the combination of diet
and exercise appears to have more effect in the delaying or preventing the
progression from IGT to a diagnosis of diabetes.” (Jones et al., 2011, p. 52) In
comparison with pharmacological interventions the authors stated that “…lifestyle
interventions seem to be at least as effective as pharmacological interventions.”
and ”…incur fewer and less serious side effects than drug treatment.” (Jones et al.,
2011, p. 14). When considering the sustainability of effects Jones et al. noted that
short-term interventions showed greater effects than medium-term interventions.
According to the authors possible explanations could be a reduction in compliance
in the intervention group with time and, in addition, individuals in placebo groups
might take up alternative strategies for dealing with their risk and might not stick to
their original lifestyle habits (Jones et al., 2011). They concluded that “…advice on
diet and exercise needs to be regularly reinforced in order to maintain behavioural
changes.” (Jones et al., 2011, p. 14).
56
5.1.2.4 Secondary Outcomes
Secondary outcomes of interest examined by Jones et al. were BMI change,
weight change, change in blood pressure, blood glucose, waist circumference and
cholesterol. The authors investigated the possibility of performing a metaregression or a network meta-analysis but stated that too few studies reported on
the outcomes of interest to do so (Jones et al., 2011). Hence, only a description of
findings in the individual RCTs is provided in the ScHAAR Review and
summarized to evidence statements. In essence the authors concluded that in the
short-term (two to five/six years), both lifestyle interventions and pharmacological
interventions showed a greater reduction in BMI, a greater weight change, a
slightly greater reduction in systolic blood pressure, diastolic blood pressure,
fasting blood glucose, two hour glucose and waist circumference in the
intervention group than in control groups. For change in cholesterol there was
mixed evidence in the intervention groups and the control groups, respectively
(Jones et al., 2011).
5.2
Patient-relevant outcomes
The evaluation of the effect of LI on patient relevant outcomes has not been within
the scope of the ScHARR Review. However, some of the other reviews identified
by our systematic literature research and some of the RCTs included in the
reviews reported on patient relevant outcomes. The basic results will be
summarized here.
5.2.1 CVD events, CVD mortality and all-cause mortality
In their review on the effectiveness of LI in patients with metabolic syndrome
Dunkley et al. stated that LI and pharmacological interventions can reverse
metabolic syndrome, but that it remained unclear whether or not these benefits
could be sustained and would translate into longer term prevention of CVD
(Dunkley et al., 2012). Sumamo et al. found no significant differences for CVD
events and mortality in 2 studies included in their review (DPS und Da Qing)
(Sumamo et. al, 2011). In addition LeBlanc et al. stated, that behaviorally based LI
are safe and effective for weight loss and maintenance and can reduce diabetes
incidence. However, data on long-term health outcomes of weight loss
interventions on CVD or death were classified insufficient by the review authors
(LeBlanc et al., 2011).
57
As mentioned by Sumamo et al. two RCTs reported on long-term outcomes on
CVD risk and mortality. Data on cardiovascular morbidity and mortality has been
presented in a secondary analysis of the Finnish DPS. After a median follow-up
time of 10.2 years, CVD morbidity (incidence rates of 22.9 and 22.0 per 1000
person-years (HR=1.04, 95% CI: 0.72-1.51 IG compared to CG)) and total
mortality rates did not differ between intervention and control group (Uusitupa et
al., 2009). The study authors considered different explanations for their findings.
First they pointed out that the DPS initially had not been designed to investigate
endpoints and therefore might lack statistical power to detect small differences
between intervention and control group. Secondly they considered that lifestyle
changes in the IG might have been extensive enough to prevent T2DM but not
extensive enough to lower CVD risk or total mortality. Thirdly they explained that
the DPS cohort had a more favorable risk profile compared to a population cohort
(FINRISK) and thus represented “…the healthier proportion of people with IGT”
(Uusitupa et al., 2009, p. 7), which could explain why no differences in CVD
morbidity or total mortality could be observed (Uusitupa et al., 2009).
In line with the DPS findings, the long term follow-up of the Da Qing study also did
not find significant differences between LI groups and control groups concerning
CVD morbidity, CVD mortality or all-cause mortality. The authors noted: “The
incidence of first CVD events and all cause mortality did not differ significantly
between the combined intervention group and the control group. The overall
adjusted HRR of death from CVD, however, was 17% lower in the intervention
group, but 95% CIs were wide and the difference was not significant. The
observation that each of the outcomes (all-cause mortality, CVD mortality, CVD
incidence) seem to have a reduced incidence for people who had received the
intervention is encouraging. Nevertheless, our findings leave the relation between
lifestyle-based diabetes prevention and effect on CVD and mortality unresolved.”
(Li et al., 2008, p. 1787).
Moreover, in a 3-year follow-up of the DPP no differences in all-cause mortality
between the three treatment groups have been observed (Knowler et al., 2002).
Although the lifestyle intervention group showed improvements in CVD risk factor
status compared with placebo and metformin therapy, no differences in CVD event
rates could be observed after 3 years (Ratner et al., 2005).
58
5.2.2 Microvascular complications
Impact of LI on microvascular complications or HRQoL have not been reported in
either the systematic reviews or in the original papers of the included RCTs.
However, secondary analyses of long term outcome data of DPS, DPP and Da
Qing study participants report on retinopathy and HRQoL.
In a 20 year follow-up of the original Da Qing study Gong et al. reported that “…the
cumulative incidence of severe retinopathy was 9.2% in the combined intervention
group and 16.2% in the control group (p=0.03, logrank test). After adjusting for
clinic and age, the incidence of severe retinopathy was 47% lower in the
intervention group than the control group (hazard rate ratio 0.53, 95% CI 0.29–
0.99, p=0.048).” (Gong et al., 2011, p. 300). Interestingly enough all the
individuals, who had been diagnosed with severe retinopathy, had also developed
diabetes. The authors concluded that the main reason for the lower incidence of
retinopathy in the intervention groups was the lower incidence of T2DM and the
associated delay of manifestation of T2DM in the IG. For nephropathy and
neuropathy no significant differences in the incidence between IG and CG could be
found (Gong et al., 2011).
In a subgroup of the DPP study population Nathan et al. compared the prevalence
of diabetic retinopathy in persons with recently diagnosed T2DM (mean duration of
T2DM 3.1 years) to the prevalence of diabetic retinopathy in persons without
T2DM but with elevated FPG and IGT. They found that retinopathy is also present
in persons with elevated FPG and IGT, but prevalence is substantially higher in
those individuals who had developed T2DM during the course of the DPP (Nathan
et al., 2007). “Retinopathy consistent with diabetic retinopathy was detected in
12.6 and 7.9% of the diabetic and non-diabetic participants, respectively (P=0.03,
comparing prevalence in the two groups).” (Nathan et al., 2007, p. 137)
5.2.3 Health Related Quality of Life (HRQoL)
Health related quality of life has been assessed in a sample of 3210 study
participants in the DPP. It was measured with the short-form health-related quality
of life (SF-36) survey at enrollment and annually thereafter (Marrero et al., 2013).
The results showed that those who remained diabetes-free had a higher score in
HRQoL no matter to which treatment arm they belonged. Diagnosis of T2DM led to
59
a marked decrease in HRQoL in all groups and did not return to baseline scores
thereafter.
In addition Florez et al. tested the hypothesis that LI in the DPP would lead to a
better HRQoL compared to the placebo intervention or pharmacotherapy with
metformin. They found that “…during the first year of intervention, physical
function, general health and vitality scores improved significantly in ILS
participants, reaching MID when compared to those treated with PLB or MET.”
(Florez et al., 2012, p. 1596) (ILS = intensive Lifestyle intervention, MID =minimal
important difference; PLB = Placebo; MET= metformin treatment) Furthermore,
HRQoL worsened in all three study groups during DPP follow-up, but the decline
was slower in the LI group. The authors concluded that lifestyle interventions in the
DPP that resulted in weight loss and increased physical activity lead to a modest
improvement of most physical HRQoL and vitality scores in overweight or obese
individuals at high risk of developing T2DM (Florez et al., 2012).
Decline of HRQoL in all DPP groups during follow-up was also reported by Marrero
et al. As one possible explanation the authors pointed out, that HRQoL declined
with age anyway (Marrero et al., 2013).
5.3
Results of interest for EBHI
In this chapter selected results from RCTs on diabetes prevention or additional
systematic reviews, which might be of special interest to patients and citizens and
thus should possibly be included in an EBHI, will be presented.
5.3.1 Adherence to lifestyle change and dose-response relationship
The results presented so far compared the change in diabetes incidence between
intervention and control groups and not between individuals in the different groups.
Consequently the effects observed did not measure the effect of implemented
lifestyle change, but to a certain degree measured how successful intervention
programs were in helping participants to change their habits. Although it is
important to know that changing lifestyle and maintaining changes is usually not
achieved by the majority of participants, from the perspective of an individual
affected it might also be interesting what the effects of implemented lifestyle
change are.
60
Data from the Finnish DPS have shown that about one third of the participants in
the lifestyle intervention group reached none or only one of the predefined targets
of lifestyle change at the year one examination. On the other hand a certain
number of participants in the control group also changed their lifestyle without
having intensive advice (Lindström et al., 2006). Therefore Lindström et al.
calculated success scores depending on the number of predefined targets reached
(0 to 5) and compared incidence rates between the different groups. They found a
strong inverse relationship between success score and diabetes incidence during
total follow-up of the DPS. “Incidence rate per 100 person-years ranged from 8.4
(95% CI 6.2–11.3) in the participants who did not achieve any of the goals at the 3year visit, to 2.0 (1·0–4·3) in those who achieved four or five of the goals. The
hazard ratios were 1·00, 0·85 (0·57–1·28), 0.66 (0·40–1·09), 0·69 (0·38–1·26),
and 0·23 (0·10–0·52) for success score from 0, 1, 2, 3, to 4–5, respectively (test
for trend p=0·0004).” (Lindström et al., 2006, p. 1676). In addition Lindström et al.
found that most people who had maintained their lifestyle change at the 3-yearvisit remained diabetes-free during the follow-up period of 7 years (Lindström et
al., 2006).
The authors concluded that “…the true effect of healthy lifestyle results in a
dramatically better outcome than that seen by the intention-to-treat analysis of the
treatment effect.” (Lindström et al., 2006, p. 1677).
However, changing lifestyle habits is a problem for a substantial number of
individuals with pre-diabetes and a challenge for diabetes prevention programs
relying on lifestyle change (Gillett et al., 2012 and Lindström et al., 2006). Gillett et
al. concluded that “…even among the volunteers in the trials, many did not
succeed and others succeeded in the short term (such as the first 6 months) but
not in the longer term. The key to success is sustained lifestyle change, especially
weight loss. In conclusion, lifestyle measures can be highly effective in reducing
progression to diabetes but adherence to lifestyle change is the most important
factor.” (Gillett et al., 2012, p. 69).
61
5.3.2 Quality criteria of effective interventions
The question of implementation of and adherence to lifestyle change is closely
linked to the question what characterizes interventions which enable participants to
change dietary intake and physical activity habits.
For individuals affected it would be helpful if an EBHI on prevention of T2DM
through LI would also provide information about how different intervention
components are linked to success in changing behavior of participants. With this
information at hand patients could ask for intervention programs in their health
care system which would help them to achieve their goals.
Two systematic reviews of reviews tried to identify intervention components
causing or being associated with increased effectiveness of lifestyle changes in
individuals with increased risk for T2DM. Greaves et al. reviewed the literature for
the IMAGE study group and Johnson et al. for the NICE (Greaves et al., 2011;
Johnson et al., 2011).
Both reviews identified the following intervention components to be causally linked
to intervention effectiveness:
•
Using well-defined/established behavior change techniques,
•
organizing or encouraging social support and
•
aiming at changes in both diet and physical activity. (Greaves et al., 2011;
Johnson et al., 2011).
In addition increased effectiveness “…was also associated with increased contact
frequency and using a specific cluster of “self-regulatory” behavior change
techniques (e.g. goal-setting, self-monitoring).” (Greaves et al., 2011, p. 119). For
other intervention components like setting, delivery mode or provider and study
population no clear relationships could be detected. Greaves et al. also stressed
the importance of implementing behavior maintenance strategies into the
programs in order to be successful in the long run (Greaves et al., 2011).
In summary both reviews concluded that intervention components, which increase
effectiveness, should be used when designing and implementing LI programs
(Greaves et al., 2011; Johnson et al., 2011).
62
5.3.3 Absolute Risk Reduction and long term progression rates to T2DM
As pointed out in chapter 3.2 it is strongly recommended that changes in risk
should be communicated as absolute risk reduction, because the exclusive
representation of relative risk reduction does not allow for the distinction of large
effects from small effects. In order to do so base rates must be available. The
progression rates from IGT to overt T2DM differ substantially between the different
population groups in the RCTs included in the ScHARR review. Table 13 shows
progression rates from the larger trials calculated by Gillett et al., 2012.
Table 13: Progression to diabetes and regression to NGT in larger trials (Gillett et
al., 2012, p. 68)
Data for the German population could not be found in the literature therefore the
calculation of absolute risk will be based on the base rate of the largest trial – the
US DPP with a 10-year annual incidence of 11 (9.3 – 13.3). Assuming a HR of
0.51 (95% CI 0.43 - 0.62) as calculated in the meta-analysis of the ScHARR
review, the calculated absolute risk in the intervention group would be 5.61 (95%
CI 4.73 - 6.82). In plain words this would mean that without lifestyle intervention 11
of 100 persons would have a diabetes diagnosis each year, with lifestyle
intervention the number would be reduced to about 5 to 6 of 100 persons per year.
One can assume that patients might also be interested in longer term outcomes
and not just annual progression rates. The DPPOS provided data on cumulative
incidence of T2DM in the different treatment groups which are shown in figure 4.
63
Year since DPP randomization
Figure 4: Cumulative incidence of diabetes in the DPP from randomization to year
10 all participants (Diabetes Prevention Program Research Group, 2009,
p. 1682)
Figure 4 shows that the cumulative incidence in the placebo group exceeds 50%
after 10 years, is about 47% in the metformin group and lower (about 40%) in the
original lifestyle intervention group. It should be noted that due to the benefits seen
in the DPP for the LI group, for ethical reasons participants of all intervention
groups were offered group-implemented lifestyle intervention after unmasking the
original assignment and starting the DPPOS. Nevertheless diabetes incidence
“…was reduced by 34% (24–42) in the lifestyle group and 18% (7–28) in the
metformin group compared with placebo” (DPPOS, 2009, p. 1677 ) 10 years after
DPP randomization. In line with the numbers presented in the DPPOS data from
the Finnish DPS showed that “…around 50% of people with impaired glucose
tolerance will develop diabetes during 10 years when no active intervention is
applied.” (Lindström et al., 2006, p. 1678). In addition the DPPOS showed that
“…onset of diabetes was delayed about 4 years by lifestyle intervention and 2
years by metformin compared with placebo” (DPPOS, 2009, p. 1683).
64
6
Discussion
The main objective of this master thesis has been the evaluation of the
effectiveness of LI in the prevention of T2DM in high-risk individuals with the
methods of EBM in order to inform EBHI which is supposed to be written by the
IQWIG Department of Health Information in the near future. Three systematic
reviews and one meta-analysis have been identified, which consistently show
that the progression to T2DM in high-risk individuals can be delayed or
prevented with LI. In addition and in contrast with the IQWIG method I have
also looked for evidence from study types other than systematic reviews of
controlled trials concerning reduction of CVD risk, mortality, microvascular
complications, HRQoL and other aspects that might be of interest to individuals
reading EBHI on diabetes prevention.
In this chapter the results which have been presented in chapter 5 will be
compared with previous reviews and with guidelines for prevention of T2DM
from
relevant
organizations.
Moreover,
methodological
considerations
concerning the information basis for EBHI will be made. Finally the relative
importance of EBHI on diabetes prevention at the population level will be
discussed from a public health perspective.
6.1
Comparison of results with previous reviews
According to our search strategy only systematic reviews, in which the
literature search had been performed 2009 or later, have been included.
However, previous systematic reviews and meta-analyses came to similar
conclusions as the reviews identified by our search. The AHRQ published a
systematic review in 2005 on the diagnosis, prognosis and treatment of IGT
and IFG and included 4 trials of combined diet and exercise interventions into a
meta-analysis. A relative risk of 0.54 (95% CI 0.42-0.70) was calculated
comparing the intervention to the control group (Santaguida et al., 2005), which
is in good agreement with the results of the ScHARR review by Jones et al. from
2011. Another meta-analysis was carried out by Gillies et al. in 2007 and found a
HR of 0.49 (95% CI 0.40 -0.59) for combined diet and exercise interventions and a
HR of 0.51 (95% CI 0.44 -0.60) when studies on all intervention types (diet-only,
exercise-only, diet & exercise) were pooled (GiIlies et al., 2007). Other systematic
reviews also came to the conclusion, that progression from IGT to T2DM can be
65
delayed or prevented by LI and that combined diet and exercise interventions are
more effective than diet-only or exercise-only interventions (Diabetes Australia,
2008 and Orozco et al., 2008).
6.2
Comparison with guidelines of relevant organizations
Evidence based guidelines for the treatment and prevention of T2DM are provided
by different organizations. The basic recommendations concerning T2DM
prevention through LI published by the American Diabetes Association and a
European study group – the IMAGE Study Group – will be summarized and
compared to the results of the ScHARR review by Jones et al. from 2011. In
addition interesting aspects of a public health guidance prepared by the NICE on
the basis of the ScHARR Review by Jones et al. and other sources of information
will be discussed.
Table 14 provides information about the guideline recommendations, the
methodology of evidence search and - where applicable - the evidence levels of
the given recommendations.
In line with the evidence provided by the ScHARR Review by Jones et al. that LI
can delay or prevent the progression to T2DM in high-risk individuals, all
organizations recommend that those affected should be offered lifestyle change
programs, which will help them to increase physical activity, change dietary habits
and lose weight. A sustained weight loss of (5 to) 7% is considered desirable and
sufficient to substantially lower the risk of T2DM by the ADA and the IMAGE study
group (both evidence level A).
Both organizations also provide recommendations concerning the amount of
physical activity and the adequate dietary change. ADA recommends at least 150
min/week of moderate activity such as walking, which corresponds with the targets
set in the DPP. The recommendation of the IMAGE study group of 30 minutes per
day of moderate exercise is in a similar range.
The IMAGE study group recommends a high-fibre, moderate fat diet with a
reduced amount of saturated and trans-fat and ADA adds the recommendation to
reduce calorie intake and to limit intake of sugar-sweetened beverages. The
recommendations of the NICE public health guidance are similar.
66
Table 14: Overview of recommendations from ADA, IMAGE Study Group and NICE
Organization/
Guideline
American Diabetes Association (ADA) 2013
Standards of Medical Care in Diabetes - 2013
Source: American Diabetes Association (ADA) (2013). Standards of Medical Care in Diabetes - 2013, Diabetes Care, Volume
36, Supplement 1, S11-66, January 2013
Evidence level
Recommendations
• Patients with IGT, IFG, or an A1C of 5.7–6.4% should be referred to an effective ongoing support
program targeting weight loss of 7% of body weight and increasing physical activity to at least 150
min/week of moderate activity such as walking.
• Among individuals at high risk for developing type 2 diabetes, structured programs that
emphasize lifestyle changes that include moderate weight loss (7% body weight) and regular
physical activity (150 min/week), with dietary strategies including reduced calories and reduced
intake of dietary fat, can reduce the risk for developing diabetes and are therefore recommended.
• Individuals at risk for type 2 diabetes should be encouraged to achieve the U.S. Department of
Agriculture (USDA) recommendation for dietary fiber (14 g fiber/1,000 kcal) and foods containing
whole grains (one-half of grain intake).
• Individuals at risk for type 2 diabetes should be encouraged to limit their intake of sugarsweetened beverages(SSBs).
• Follow-up counseling appears to be important for success.
(IGT) A
IFG(E)
A1C of 5.7–6.4% (E)
• Metformin therapy for prevention of type 2 diabetes may be considered in those with IGT, IFG, or
an A1C of 5.7–6.4% (E), especially for those with BMI >.35 kg/m2, aged < 60 years, and women
with prior GDM.
IGT (A)
IFG(E)
A1C of 5.7–6.4% (E)
BMI >.35 kg/m2 +
aged < 60 years, and
women with prior
GDM. (A)
B
• Screening for and treatment of modifiable risk factors for CVD is suggested.
A
B
B
Evidence search and
included studies
/RCTs on lifestyle
intervention
ADA Professional Practice Committee members systematically searched Medline for human studies published since 1 January
2011. Recommendations were revised based on new evidence or, in some cases, to clarify the prior recommendation or match
the strength of the wording to the strength of the evidence. A table linking the changes in recommendations to new evidence can
be reviewed at http://professional.diabetes.org/CPR.
Grading system
Included studies: DPP, DPS, Da Qing, IDPP
ADA evidence grading system for clinical practice recommendations (details see appendix 4)
67
Table 14: Overview of recommendations from ADA, IMAGE Study Group and NICE (con`t)
Organization/
Guideline
IMAGE Study Group, Paulweber et al., 2010
A European Evidence-Based Guideline for the Prevention of Type 2 Diabetes
Source: Paulweber, B. et al., (2010). A European Evidence-Based Guideline for the Prevention of Type 2 Diabetes. Hormone and
Metabolic Research, Volume 42, page S1–S64, April 2010
Recommendations
(A+B level Evidence)
• Intensive lifestyle interventions that encourage people to change their diet and to increase their level of
physical activity should be used to prevent or delay the onset of T2DM in adults with IGT. The NNT for
prevention of one case of T2DM of 6.4 [95% CI 5.0, 8.4] at mean follow up ranging from 1.83 to 4.62 years.
• Weight reduction is an essential element of T2DM prevention. Sustained weight reduction by 5–7% is
sufficient to substantially lower the risk of T2DM.
• An increase in physical activity even at a level of 30 minutes per day of moderate exercise reduces the risk
of T2DM and is therefore recommended.
Evidence
level
A
A
B
Evidence search and
included studies
/RCTs on lifestyle
intervention
B
• A diet with high fibre (≥ 15 g per 1000 kcal), moderate fat (≤ 35% of total energy) reduced saturated and
trans fat (< 10% of total energy) can lower body weight and reduce the risk of T2DM and is therefore
recommended.
• Interventions should
A
• aim to promote changes in both diet and physical activity
A
• use established, well defined behavior change techniques (e. g., specific goal-setting, relapse
A
prevention, self-monitoring, motivational interviewing).
A
• encourage participants to engage social support for the planned behaviour change
A
• include a strong focus on maintenance.
B
• maximize the frequency of number of contacts within the resources available.
• systematic search for primary studies, systematic reviews and meta-analyses of research on preventing the onset of T2DM.
• initial search was undertaken using MEDLINE with follow-up of cited references.
• final selection limited to randomized controlled trials (RCTs) published in English between1979 and 2008, which featured
development of T2DM as a study endpoint and used standard criteria for the diagnosis of diabetes mellitus.
Grading system
Included studies: DPP, DPS, Da Qing, IDPP, Japanese trial in IGT males (Kosaka et al., 2005 )
SIGN (details see appendix 4)
68
Table 14: Overview of recommendations from ADA, IMAGE Study Group and NICE (con`t)
Organization/
Guideline
National Institute for Health and Care Excellence (NICE) (2012). Preventing type 2 diabetes: risk identification
and interventions for individuals at high risk. (NICE public health guidance 38)
Source: National Institute for Health and Care Excellence (NICE) (2012). Preventing type 2 diabetes: risk identification and
Evidence level
interventions for individuals at high risk. NICE public health guidance 38, retrieved from
http://www.nice.org.uk/nicemedia/live/13791/59951/59951.pdf on 18th of November 2013
Recommendations
• For people confirmed as being at high risk a referral to a local, evidence-based, quality-assured intensive
lifestyle change program should be offered.
• Intensive lifestyle-change programs should offer ongoing tailored advice, support and encouragement to
help people:
• undertake a minimum of 150 minutes of 'moderate-intensity' physical activity per week
• gradually lose weight to reach and maintain a BMI within the healthy range
• increase their consumption of whole grains, vegetables and other foods that are high in dietary fibre
• reduce the total amount of fat in their diet
• eat less saturated fat.
• Established behaviour-change techniques should be used, inter alia information provision, exploration
and reinforcement of participants' reasons for wanting to change, goal setting, action planning, coping
plans and relapse prevention
• Participants should be encouraged to involve a family member, friend or carer who can offer emotional,
information, planning or other practical support to help them make the necessary changes.
• Participants should be encouraged to use self-regulation techniques.
Evidence search and
included studies
/RCTs on lifestyle
intervention
Grading system
• Evidence relies on the ScHAAR review and meta-analysis by Jones et al., 2011, which has been described in detail in chapter
5.1.2 and on additional systematic reviews and expert papers which can be found on http://publications.nice.org.uk/preventingtype-2-diabetes-risk-identification-and-interventions-for-individuals-at-high-risk-ph38/appendix-e-supporting-documents
-
69
Interestingly enough none of the organizations raises the question, whether
prevention of T2DM should be considered a surrogate parameter only or if T2DM
is a valid surrogate for CVD or mortality. Furthermore, the organizations do not
discuss if hard endpoints like CVD risk or mortality reduction should be proven in
order to recommend lifestyle interventions to patients and health care providers.
One reason could be that the evidence base of the reviews, on which the
recommendations are based, is broader and also included study types other than
RCTs.
As pointed out in chapter 5.2 significant reductions in long term adverse health
outcomes like CVD risk, CVD or all-cause mortality could neither be found in the 4
identified reviews nor in the individual RCTs that have been included in the
ScHARR review by Jones et al. However, risk calculations from a populationbased reference study cohort (FINRISK) of the Finnish DPS show that all-cause
mortality in people with IGT is higher than in those with NGT, but lower than allcause mortality rates of people with known T2DM. The mortality rates were 6.6,
16.4, 21.0, and 28.8 per 1000 person-years in the NGT, IGT, ST2DM and known
T2DM groups, respectively. The adjusted HRs for all-cause mortality were 0.52
(0.36–0.74) and 1.96 (1.15–3.34) for NGT and known T2DM compared to IGT,
respectively (Uusitupa et al., 2009). Furthermore, there was also a statistically
significant - albeit marginal - difference in CVD event rates between individuals
with IGT compared to those with known T2DM (adjusted HR for known T2DM
compared to IGT: 1.64 (1.02–2.15) (Uusitupa et al., 2009).
In addition Kowall et al. investigated all-cause and cause-specific mortality in an
older German population, a total of 1466 subjects aged 55-74 years from the
KORA population based survey (Kowall et al., 2011). They found that the age and
sex adjusted mortality rate for 1000 person years was substantially higher in
individuals with undiagnosed or known T2DM compared to individuals with IGT
(IGT: 10.52 (7.67–14.42), undiagnosed T2DM: 33.05 (21.88–49.91), known T2DM:
28.78 (20.62–40.16)). Kowall et al. concluded that “… all-cause mortality in
persons with undiagnosed or known diabetes is strongly increased compared with
persons with prediabetes and NGT.” (Kowall et al., 2011, p. 643)
In conclusion there is evidence from population-based observational studies but
not from RCTs in high-risk individuals that mortality rates are lower in people with
IGT compared to individuals with T2DM. In addition the likelihood of remaining
70
diabetes free or returning to NGT is greater for individuals in the IGs than those in
the CGs in all included reviews or large RCTs, respectively (see figure 4) (Gillett et
al., 2012).
There are several possible explanations why no significant risk reductions for CVD
or mortality could be observed in the lifestyle groups compared to the control
groups. As Uusitupa et al. pointed out the Finnish DPS - and most of the other
trials - had initially not been designed to investigate endpoints and therefore might
lack statistical power to detect small differences (Uusitupa et al., 2009).
Furthermore, it is well known that volunteers in randomized controlled trials usually
are not representative for the whole population, because they tend to be more
health-conscious, motivated and educated than the general population. This also
applies for the control group and has been shown for the Finnish DPS (Uusitupa et
al., 2009). In addition data from the Finnish DPS has revealed that a certain
number of participants in the control group also changed lifestyle habits over the
years. At first post-intervention follow-up visit 40% of the control group participants
had achieved one of the lifestyle goals and 7% had achieved at least four out of
five predefined targets (Lindström et al., 2006). This kind of “contamination” in the
control group could be one additional reason for the lack of significant differences
in CVD risk and mortality between control and intervention group in the trials.
Besides those methodological considerations it cannot be ruled out that lifestyle
interventions in high-risk individuals simply are not able to reduce CVD risk and
mortality. Reasons could be that the implemented lifestyle changes might be
sufficient to reduce T2DM incidence but not intensive enough to reduce CVD event
rates and mortality. Another reason could be that lifestyle change came too late to
reverse the deleterious effects of overweight, lack of physical activity and
unhealthy diet which presumably have acted upon the organism of most high-risk
participants for years. As pointed out in the Joint ESC Guidelines from 2012
evidence has increased over the last decades that cardiovascular disease risk
starts to develop at a young age and thus a healthy lifestyle in the young is crucial
to prevent CVD events (European Society of Cardiology, 2012). In addition
Lindström et al. concluded that “The high diabetes incidence even in the
intervention group of our study suggests that preventive actions should probably
be targeted to all high-risk individuals, even before impaired glucose tolerance is
present.” (Lindström et al., 2006, p. 1678)
71
Last but not least cardiovascular disease has a multi-factorial genesis (European
Society of Cardiology, 2012) and it can be assumed that many of the trial
participants had multiple risk factors in addition to IGT since many of them had
been diagnosed with metabolic syndrome, at least in the IDPP, the DDP, the DPS
and the Da Qing trial. Therefore other risk factors like elevated blood pressure,
blood cholesterol levels and abdominal obesity might be responsible for the fact
that improved glycemic control or prevention of T2DM did not translate into
reduced CVD event rates or reduced mortality.
6.3
Methodological considerations
This chapter deals with methodological considerations. First of all it will be
discussed if systematic reviews and meta-analyses are the appropriate method for
the evaluation of complex interventions. Secondly the question will be raised
whether relying solely on reviews of controlled studies as information source for
EBHI is appropriate with respect to the patients’ or citizens’ information needs.
From a methodological point of view lifestyle intervention programs should be
considered as complex interventions. Complex interventions are defined by
Mühlhauser et al. as follows: “They comprise interdependent components
differently interacting within various complex settings.” (Mühlhauser et al., 2011, p.
752).
Applying this definition to lifestyle interventions for the prevention of T2DM,
interdependent intervention components would be, for example, a change in
dietary fat intake, calorie intake and weight loss. Furthermore, the intervention
components used to deliver the program can also be regarded as complex: the
use or selection of behavior change strategies, skills of program delivery personnel
or the delivery mode are examples of program characteristics which are likely to
interact and distinguish different programs from each other. In addition the
interventions would typically address different target groups in different settings
(e.g. people of Asian ethnicity in an Indian, Chinese, Japanese or US cultural
setting compared to Finnish participants in a European cultural setting).
Furthermore, the trial participants are part of social systems in their families or at
their workplace which influence success in lifestyle change and it can be assumed
that the influences differ substantially between the different settings.
72
Although there is “…no sharp boundary between simple and complex
interventions” (Craig et al., 2008) it is obvious that the success of changing eating
behavior or physical activity patterns depends on more influencing factors than for
example drug treatment and thus can be regarded as complex. This might have
consequences concerning the appropriate method of evaluation.
Mühlhauser et al. pointed out that “….Appraising the efficacy, benefit and harm of
complex interventions is far more difficult than appraising single interventions like
specific drug treatments.” (Mühlhauser et al., 2011, p. 752). In addition Lenz et al.
argued that meta-analysis might not be the appropriate method to appraise
complex interventions and that “…pooling of outcome measures across different
programmes is usually inappropriate” (Lenz et al., 2007, p. 1375). After reviewing
the literature and methodological guidance of international organizations on the
appraisal and synthesis of complex interventions they concluded: “Complex
interventions require multistage development, use of different methods, reporting
on all developing phases and new approaches for synthesis. Presentation of the
complete evidence on a specific complex intervention might be more useful than
synthesis of a variety of different complex interventions by customarily applied
methods of (metaanalytical) systematic review.” (Mühlhauser et al., 2011, p. 752)
Applying this line of reasoning to the topic of this master thesis it is worth
discussing if systematic reviews or meta-analyses are the (only) appropriate
method of evaluating effectiveness of lifestyle interventions or gaining information
for EBHI.
First of all it should be considered that the included RCTs in the ScHARR review
showed substantial differences concerning intervention type, intensity and
duration, follow-up time, recruits in terms of ethnicity, age, BMI and other
characteristics as well as study quality. Nevertheless all RCTs were included in the
meta-analysis. In contrast Gillett et al. pointed out that in their review from 2012
meta-analysis was not performed because of these differences (Gillett et al.,
2012).
Secondly, one could argue that valuable information provided by single RCTs is
lost, when only evidence from systematic reviews is considered appropriate. As
illustrated in chapter 5.2 there is evidence on the basis of RCTs that lifestyle
interventions provide benefits in terms of development of retinopathy and HRQoL
73
in high-risk individuals. In addition results of population based observational
studies indicate that mortality rates are lower in people with IGT compared to
individuals with T2DM. Accordingly as pointed out in chapter 6.2 all organizations,
which prepared evidence based guidelines on diabetes prevention, did not
question the view that delay or prevention of T2DM is a benefit to patients.
In addition a wider use of all evidence and not just evidence from systematic
reviews of controlled trials may be justified by the information needs of consumers
and patients. In a paper of the European Medicines Agency, which deals with the
expectations of patients, consumers and healthcare professionals, the authors
pointed out that “…a great deal of information on the benefits and risks of
medicines is expected; qualitative and quantitative descriptions of the benefits and
risks, data in specific sub-populations, comparative data with alternative
treatments, success factors and risk factors.” (EMEA, 2009) Relating this general
description of information needs to the topic of this master thesis it became
obvious that in depth information on adherence, quality of interventions, doseeffect-relationships and long term outcomes has not been available on the level of
systematic reviews but in contrast is available from individual RCTs (see chapter
5.3).
The main argument for relying mainly on systematic reviews of RCTs from the
IQWIG point of view is the higher certainty of findings and statements. On the
other hand the concept of health literacy comprises also the consumers`
knowledge about the inherent element of uncertainty in scientific findings.
Therefore EBHI could also aim to raise awareness of uncertainty in medical
knowledge by presenting knowledge on different evidence levels rather than only
giving information which has been proven on the level of systematic reviews.
It should therefore be considered whether a system, which is similar to the levels
of evidence in evidence based guidelines, could be developed, which would inform
patients and consumers in an easily understandable way about the certainty of the
communicated information. This would make it more likely that statements could
be made in EBHI about issues, which are relevant to patients and consumers.
However, it is well known that the knowledge about different study types and their
ability to provide evidence is very limited in the general population (Klemperer and
Dierks, 2012). Therefore close collaboration with users and evaluation of
understanding of such an information system would be desirable.
74
With respect to evidence levels and study types it should be considered that the
Oxford Centre of Evidence Based Medicine in its publication on “Levels on
Evidence” from March 2009 also regards individual RCTs with narrow confidence
intervals as level 1b evidence which could translate into a grade A
recommendation (see appendix 5) (Oxford Centre for Evidence Based Medicine,
2009). In addition organizations that are primarily concerned with issues of
prevention like WHO or the World Cancer Research Fund have developed
systems of evidence synthesis that take into account evidence from study types
other than RCTs (WHO, 2003; WCRF, 2007). Summing up, the following questions
should be answered before preparing EBHI on the prevention of T2DM:
1. Should T2DM be considered a surrogate endpoint and if yes, is it a valid
surrogate marker for patient-relevant outcomes?
2. Should the term “sufficiently strong evidence” (see table 7) be translated
into evidence on the basis of systematic reviews only or should evidence
from other types of studies also be included in order to meet the
information needs of consumers and patients.
3. What are the questions of high-risk and also lower risk individuals (i.e.
consumers in general) that should be answered by EBHI?
4. What do experts in the field think about the issues discussed here?
6.4
The relative importance of EBHI on diabetes prevention from a public health
perspective.
As this master thesis has been written in order to attain the degree of a Master of
Public Health brief consideration will be given to the relative importance of EBHI
from a public health perspective.
EBHI on the effectiveness of lifestyle changes on diabetes risk may be a
necessary but not sufficient condition for behavior change for the individual. It is
well known that reading and understanding information is not the same as acting
according to that information. As Marstedt and Rosenbrock put it: “Between the
knowledge and the understanding of a health message and its transposition into
one`s own way of life, a number of hurdles (Rosenbrock and Michel 2006) appear,
which make the great majority even of those, who would like to follow the
message, fail.” (Marstedt and Rosenbrook, 2010, own translation). As experts in
the field are well aware of this problem, simply providing information on a healthy
lifestyle has been the control intervention in most of the trials. In other words, EBHI
75
on the effectiveness of lifestyle changes may generate the desire for behavioral
change in high risk individuals, but other factors may determine whether such
intentions result in an actual change of lifestyle and risk.
On the other hand it is conceivable that an evidence based statement of the
IQWIG about the effectiveness of LI on the prevention of T2DM could also reach
policy makers, change their knowledge, alter attitudes to the topic and eventually
lead to implementation of lifestyle intervention programs for high-risk individuals in
Germany.
From a broader public health perspective strategies to reduce diabetes incidence,
which focus solely on programs that aim at behavior change of the individual
affected, do not seem to be very promising. As Dahlgren and Whitehead pointed
out in 1991 already, health is determined by a variety of influences which can be
grouped into different categories as shown in figure 5. General socio-economic,
cultural and environmental living conditions as well as the material and social
conditions people live in are of major influence on health. Furthermore, the social
networks – that is support of family, friends and communities – greatly influence
health but also lifestyle habits of the individual (Dahlgren and Whitehead, 1991).
Figure 5:
The main determinants of health (Dahlgren and Whitehead, 1991,
p.11)
76
This might partially explain why so many participants in the LI programs had
difficulty making or maintaining lifestyle changes. Dahlgren and Whitehead pointed
out: “All too often, strategies are only considered at one policy level, yet concerted
effort at several levels would in many cases be far more effective. The reinforcing
– synergetic – effects of this type of vertical health policy is in fact the very key for
improving the impact of health policies in general and strategies to reduce social
inequities in particular.” (Dahlgren and Whitehead, 1991, p. 11)
Translated to the topic of this master thesis it can be concluded that a public health
approach on diabetes prevention should tackle different layers of influence on
health simultaneously, identify and possibly change the behavioral incentives in
the living environments (Marstedt and Rosenbrock, 2010), so that a healthy
lifestyle can be implemented more easily.
Elements included in a public health diabetes prevention strategy could for
example encompass legislative changes (policies to reduce specific nutrients in
foods e.g. trans fats, and to improve labeling and advertising regulations),
sustained media and educational campaigns, multi-component school and
workplace interventions, economic incentives (e.g. strategies to lower prices of
healthier foods and beverages) and community support to produce favorable
environments for physical activity (e.g. access to sport facilities and urban design)
(Paulweber et al., 2010; Mozaffarian et al., 2012). Also the WHO stated: “The
public and private sectors also have an important role to play in developing and
implementing policies and programmes that increase knowledge about diabetes,
its prevalence and consequences, encourage and provide greater opportunities for
greater physical activity, and improve the availability and accessibility of healthy
foods.” (WHO Europe, 2013)
Summing up it seems most likely that the implementation of lifestyle intervention
programs for high-risk individuals combined with comprehensive public health
policies will be the most promising strategy to reduce diabetes incidence at the
population level. This view is supported by scientific papers by NICE and the
IMAGE study group alike (NICE, 2011; Paulweber et al., 2010)
77
7
Conclusions and Outlook
The systematic search for evidence has shown that lifestyle interventions can
delay or prevent the progression to T2DM in high-risk individuals. The pooled HR
of the intervention groups versus controls was 0.51 (95% CI 0.43-0.62) in a metaanalysis of 11 randomized controlled trials. Assuming an annual diabetes
incidence of 11% as in the DPP study, the calculated absolute risk of diabetes
would be 5.61% (95% CI 4.73- 6.82) in the intervention groups. Thus it can be
concluded that the systematic search for the best available external evidence
resulted in a suitable basis for the production of EBHI on diabetes prevention.
These results are, however, restricted to a very specific population, namely highrisk individuals. However, results from epidemiological studies in the general
population also show that individuals adhering to a healthy lifestyle combining
several factors like physical activity, diet, smoking and alcohol habits can
substantially lower their risk for T2DM (Mozaffarian et al., 2009).
Evidence for the benefit of LI in terms of patient relevant outcomes like CVD risk,
CVD mortality, all-cause mortality, morbidity or HRQoL in high-risk individuals is
less clear. As pointed out in chapter 5.2 significant reductions in long term adverse
health outcomes could neither be found in the 4 identified reviews nor in the
individual RCTs that have been included in the ScHARR review. However, long
term follow-up data of individual RCTs indicated benefits of LI in terms of
microvascular complications like diabetic retinopathy and HRQoL. Open questions,
which should be addressed by future research, therefore relate to the optimal time
of intervention, the required intensity of lifestyle change and above all, how
adherence to and maintenance of lifestyle change can be supported. Moreover, it
will be a task for the future to translate the success from randomized controlled
trials to everyday patient care in health care systems.
Methodological considerations lead to the question, whether or not EBHI on
preventive issues should solely rely on the results of systematic reviews of
controlled trials. Although this systematic literature research retrieved enough highlevel, external evidence on the preventive effects of LI in high-risk individuals, it
can be doubted that a similar search for evidence focusing on the general
population would yield enough evidence on the basis of reviews of CTs or even
individual RCTs, simply because such studies are unlikely to be feasible in practice
nor fundable. Therefore EBHI on preventive issues relying solely on high-level
78
evidence is in danger of being flawed by inherent methodological problems of
studies in primary prevention and issues of research agenda bias. Consequently, it
should be considered to develop a communication tool similar to the evidence
levels used in guidelines that would help consumers and patients understand the
different levels of certainty/evidence in scientific knowledge. This would probably
allow for communicating also findings from well-designed cohort studies,
describing detailed information from individual RCTs, disseminating EBHI on a
broader range of topics and thus meet the information needs of the general
population concerning questions in primary prevention. In addition, once
understood patients and consumers themselves could decide which level of
evidence they consider sufficient or necessary for their decisions, which would
strengthen patient and consumer autonomy.
For the future it will be very interesting in which way the IQWIG Department of
Health Information will develop their methods for the preparation of EBHI on issues
of primary prevention.
79
References:
American Diabetes Association (ADA) (2003). Follow-up Report on the Diagnosis of
Diabetes Mellitus. Diabetes Care, 26 (11), 3160-3167
American Diabetes Association (ADA) (2013). Standards of Medical Care in Diabetes –
2013. Diabetes Care, 36, Supplement 1, S11-66
Barr, E.L.M., Zimmet, P.Z., Welborn,T.A., Jolley, D., Magliano, D.J., Dunstan, D.W.,
Cameron, A.J., Dwyer, T., Taylor, H.R., Tonkin, A.M., Wong, T.Y., McNeil, J., Shaw, J.E.
(2007). Risk of Cardiovascular and All-Cause Mortality in Individuals With Diabetes
Mellitus, Impaired Fasting Glucose and Impaired Glucose Tolerance: The Australian
Diabetes, Obesity and Lifestyle Study (AusDiab). Circulation, 116, 151-157
Bo, S., Ciccone, G., Baldi, C., Dusio, F., Forastiere, G., Lucia, C., Nuti, C.,, Durazzo, M.,
Cassader, M., Gentile, L., Pagano, G. (2007). Effectiveness of a lifestyle intervention on
metabolic syndrome. A randomized controlled trial. Journal of General Internal Medicine,
22(12), 1695-1703
Bundesärztekammer
(BÄK),
Kassenärztliche
Bundesvereinigung
(KBV),
Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Fachgesellschaften (AWMF)
(2013). Nationale VersorgungsLeitlinie Therapie des Typ-2-Diabetes – Kurzfassung.
Version 1.0. 2013. Retrieved from
http://www.versorgungsleitlinien.de/themen/diabetes2/dm2_Therapie on 30th of October
2013
Bundesministerium für Gesundheit, Bundesministerium für Justiz. (2003). Patientenrechte
in Deutschland. Leitfaden für Patienten und Ärzte. Retrieved from
http://www.uke.de/studierende/downloads/zg-studierende/Patientencharta.pdf on 30th of
October 2013
Bundesministerium für Justiz.. Patientenrechte im Klartext. Retrieved from
http://www.bmj.de/DE/Buerger/gesellschaft/Patientenrechte/_node.html on 30th of October
2013
Coutinho, M., Gerstein, H.C., Wang, Y., Yusuf, S. (1999). The Relationship Between
Glucose and Incident Cardiovascular Events. A metaregression analysis of published data
from 20 studies of 95,783 individuals followed for 12.4 years. Diabetes Care, 22, 233–240
Craig, P. Dieppe, P. Macintyre, S., Mithcie, S. Nazareth, I. Petticrew, M. (2008).
Developing and evaluating complex interventions: the new Medical Research Council
guidance. British Medical Journal, 25 October 2008, 337, 979-983
80
Dahlgren, G., Whitehead, M. (1991). Policies and strategies to promote social equity in
health. Background document to WHO-Strategy paper in Europe. Institute for Future
Studies. Retrieved from
http://www.framtidsstudier.se/wpcontent/uploads/2011/01/20080109110739filmZ8UVQv2wQFShMRF6cuT.pdf on 15th of
November 2013
Diabetes Australia (2008). Evidence based guideline for the primary prevention of type 2
diabetes. Retrieved from
www.diabetesaustralia.com.au/PageFiles/763/Primarypreventionguideline.pdf on 30th of
October 2013
Diabetes Prevention Program Research Group (2006). Relationship of body size and
shape to the development of diabetes in the diabetes prevention program. Obesity, 14,
2107-17
Diabetes Prevention Program Research Group (2009). 10-year follow-up of diabetes
incidence and weight loss in the Diabetes Prevention Program Outcomes Study. The
Lancet, 374, 1677-1686.
Dunkley A.J., Charles, K., Gray, L.J., Camosso-Stefinovic, J., Davies, M.J., Khunti, K.
(2012) Effectiveness of interventions for reducing diabetes and cardiovascular disease
risk in people with metabolic syndrome: systematic review and mixed treatment
comparison meta-analysis. Diabetes, Obesity and Metabolism, 14(7), 616-625.
Emerging Risk Factor Collaboration (The) (ERFC) (2011). Diabetes mellitus, Fasting
Glucose and Risk of Cause-Specific Death. New England Journal of Medicine, 364, 82941
Eriksson, J., Lindström, J., Valle, T., Aunola, S., Hämäläinen, H., Ilanne-Parikka, P.,
Keinänen-Kiukaanniemi, S., Laakso, M., Lauhkonen, M., Lehto, P. Louheranta, A.,
Mannelin, M., Martikkala, V. Rastas, M., Sundvall, J., Turpeinen, A., Viljanen, T.,
Uusitupa, M., Tuomilehto, J. (1999). Prevention of Type II diabetes in subjects with
impaired glucose tolerance: the Diabetes Prevention Study. Study Design and one year
interim report on the feasibility of the lifestyle intervention programme. Diabetologica, 42
(7), 793-801
European Medicines Agency (2009). Information on benefit-risk of medicines: patients’,
consumers’ and healthcare professionals’ expectations. Report by the Patients’ and
Consumers’ Working Party (PCWP) and the Healthcare Professionals’ Working Group
(HCP WG). Retrieved from
http://www.ema.europa.eu/docs/en_GB/document_library/Other/2009/12/WC500018433.p
df on 14th of November 2013
81
European Society of Cardiology (2012). European Guidelines on cardiovascular disease
prevention in clinical practice (Version 2012). European Heart Journal, 33, 1635-1701
Florez, H., Pan, Q., Ackermann, R.T., Marrero, D.G., Barrett-Connor, E., Delahanty, L.,
Kriska, A., Saudek, C.D., Goldberg, R.B., Rubin, R.R. for the Diabetes Prevention
Program Research Group (2012). Impact of Lifestyle Intervention and Metformin on
Health-Related Quality of Life: the Diabetes Prevention Program Randomized Trial.
Journal of General and Internal Medicine, 27(12), 1594-1601
Gaissmaier, W., Gigerenzer, G. (2011). When Misinformed Patients Try to Make Informed
Health Decisions. In: G. Gigerenzer and J.A. Muir Gray (Hrsg.), Better Doctors, Better
Patients, Better Decisions – Envisioning Health Care 2020 (pp 29-43). Cambridge,
Massachusetts, London, England: The MIT Press
Gillies, C.L., Abrams, K.R., Lambert, P.C., Cooper, N.J., Sutton, A.J., Hsu, R.T., Khunti, K.
(2007). Pharmacological and lifestyle interventions to prevent or delay type 2 diabetes in
people with impaired glucose tolerance: systematic review and meta-analysis. British
Medical Journal, 334, 299-302
Gillett, M., Royle, P., Snaith, A., Scotland, G., Poobalan, A., Imamura, M., Black, C.,
Boroujerdi, M., Jick, S., Wyness, L., McNamee, P., Brennan, A., Waugh, N. (2012). Nonpharmacological interventions to reduce the risk of diabetes in people with impaired
glucose regulation: a systematic review and economic evaluation. Health Technology
Assessment 2012, 16(33)
Gong, Q., Gregg, W., Wang, J. An, Y., Zhang, P., Yang, W. Li, H., Jiang, Y., Shuai, Y.,
Zhang, B., Zhang, J., Gerzoff, R.B., Roglic, G., Hu, Y., Li, G., Bennett, P.H. (2011). Longterm effects of a randomised trial of a 6-year lifestyle intervention in impaired glucose
tolerance on diabetes-related microvascular complications: the China Da Qing Diabetes
Prevention Outcome Study. Diabetologia, 54, 300–307
Greaves, C.J., Sheppard, K.E., Abraham, C., Hardeman, W., Roden, M., Evans, P.H.,
Schwarz, P., The IMAGE Study Group (2011). Systematic review of reviews of
intervention components associated with increased effectiveness in dietary and physical
activity interventions. BioMed Central Public Health 2011, 11:119
Han, T.S., Sattar, N., Lean, M. (2006). ABC of obesity. Assessment of obesity and its
clinical implications. British Medical Journal, 333, 695-698
Hart, C.L., Hole, D.J., Lawlor, D.A., Davey, S.G. (2007). How many cases of type 2
diabetes mellitus are due to being overweight in middle age? Evidence from the Midspan
prospective cohort studies using mention of diabetes mellitus on hospital discharge or
death records. Diabetic Medicine, 24, 73-80
82
Heidemann C., Du Y., Scheidt-Nave, C. (2011). Diabetes mellitus in Deutschland. in:
Robert Koch-Institut Berlin (Hrsg.), GBE kompakt 2(3) www.rki.de/gbe-kompakt (Stand:
06.05.2011), Retrieved from
http://www.rki.de/DE/Content/Gesundheitsmonitoring/Gesundheitsberichterstattung/GBED
ownloadsK/2011_3_diabetes.html?nn=2531734 on 16th of August 2013
Heidemann, C., Du, Y., Schubert, I., Rathmann, W., Scheidt-Nave, C. (2013). Prevalence
and temporal trend of known diabetes mellitus. Bundesgesundheitsblatt, 56, 668-677
Institute of Medicine (IOM) (2004) Health Literacy: A prescription to end the confusion.
Report Brief, Retrieved from
http://www.iom.edu/~/media/Files/Report%20Files/2004/Health-Literacy-A-Prescriptionto-End-Confusion/healthliteracyfinal.pdf on 16th of August 2013
International Diabetes Federation (IDF) (2012). IDF Diabetes Atlas Update 2012,
Retrieved from http://www.idf.org/diabetesatlas/5e/Update2012 on 15th of August 2013
Institut für Qualität und Wirtschaftlichkeit im Gesundheitswesen (IQWIG) (2011). General
Methods Version 4.0 of 23.09.2011, Retrieved from
https://www.iqwig.de/download/IQWiG_Methoden_Version_4_0.pdf on 16th of August
2013
Jarrett, R. J., Keen, H., Fuller, J. H., McCartney, M. (1979) Worsening to Diabetes in Men
with Impaired Glucose Tolerance ("Borderline Diabetes"). Diabetologia, 16(1), 25-30.
Jones, R., Freeman, C., Johnson, M., Stevens, J., Buckley Woods, H., Guillaume, L.,
Gillies, C., Goyder, E., Chilcott, J., Payne, N., ScHARR Public Health Collaboration
Centre (2011). Preventing the progression of pre-diabetes to type 2 diabetes in adults.
Systematic review and meta-analysis of lifestyle, pharmacological and surgical
interventions. Retrieved from http://guidance.nice.org.uk/PH38/SupportingEvidence on
18th of September 2013
Johnson, M., Jones, R., Freeman, C., Buckley Woods, H., Gillett, M., Ram, V., Sidwell, A.,
Goyder, E., Chilcott, J., Payne, N. ScHARR Public Health Collaboration Centre (2011).
Prevention of type 2 diabetes: Reviewing mechanisms of successful interventions and
translation
of
major
trial
evidence
to
practice.
Retrieved
from
th
http://guidance.nice.org.uk/PH38/SupportingEvidence on 18 of September 2013
Katz, R. (2004). Biomarkers and Surrogate Markers: An FDA Perspective. NeuroRx_: The
Journal of the American Society for Experimental NeuroTherapeutics, 1, 189–195
Klemperer, D., Lang, B., Koch, K., Hilda, B., Brunsmann, F., Burckhardt, M., Dierks, M-L.,
Ehrmann, U., Günther, J., Härter, M., Mühlhauser, I., Sänger, S., Simon, D., Steckelberg,
A. (2010). Die „Gute Praxis Gesundheitsinformation“ Zeitschrift für Evidenz, Fortbildung
und Qualität im Gesundheitswesen (ZEFQ), 104, 66-68
83
Klemperer, D., Dierks, M-L. (2012). Evidenzbasierte Medizin und Qualitätssicherung
medizinischer Leistungen: Erfahrungen und Einschätzungen der Bürger. in: Jan Böcken,
Bernard Braun, Uwe Repschläger (Hrsg.). Gesundheitsmonitor 2011 (pp. 32-55).
Gütersloh, Verlag Bertelsmann Stiftung
Knowler, W.C., Barrett-Connor, E., Fowler, S.E., Hamman, R.F., Lachin, J.M., Walker,
E.A., Nathan, D.M.( 2002). Reduction in the incidence of type 2 diabetes with lifestyle
intervention or metformin. New England Journal of Medicine, 346(6), 393-403
Kosaka, K., Noda, M., Kuzuya, T. (2005) Prevention of type 2 diabetes by lifestyle
intervention: a Japanese trial in IGT males. Diabetes Research & Clinical Practice, 67(2),
152-162.
Kowall, B. Rathmann, W., Heier, M., Giani, G., Peters, A., Thorand, B., Huth, C., Icks, A.,
Meisinger, C. (2011). Categories of glucose tolerance and continuous glycemic measures
and mortality. European Journal of Epidemiology, 26, 637–645
LeBlanc, E.S., O'Connor, E., Whitlock, E.P., Patnode, C.D., Kapka, T. (2011). Screening
for and Management of Obesity and Overweight in Adults. Evidence Report No. 89.
AHRQ Publication No. 11-05159-EF-1. Rockville, MD: Agency for Healthcare Research
and Quality; October 2011
Levitan, E.B., Song, Y., Ford, E.S., Liu, S. (2005). Is nondiabetic hyperglycemia a risk
factor for cardiovascular disease? A meta-analysis of prospective studies. Journal of the
American Medical Association, 293, 194-202 cited in WHO ((2006). Definition and
diagnosis of diabetes mellitus and intermediate hyperglycemia: report of a WHO/IDF
consultation. Geneva 2006 Retrieved from
http://whqlibdoc.who.int/publications/2006/9241594934_eng.pdf on 12th of August 2013
Li, G., Zhang, P., Wang, J., Gregg, E.W., Yang, W., Gong, Q., Li, H. Li, H. Jiang, Y., An,
Y. Shuai, Y., Zhang, B. Zhang, J. Thompson, T.J., Gerzoff, R.B., Roglic, G. Hu, Y.
Bennett, P.H. (2008). The long-term effect of lifestyle interventions to prevent diabetes in
the China Da Qing Diabetes Prevention Study: a 20-year follow-up study. The Lancet,
371, 1783–89
Liao, D., Asberry, P. J., Shofer, J. B., Callahan, H., Matthys, C., Boyko, E. J., Leonetti, D.,
Kahn, S. E., Austin, M., Newell, L., Schwartz, R. S., Fujimoto, W. Y. (2002). Improvement
of BMI, body composition and body fat distribution with lifestyle modification in Japanese
Americans with impaired glucose tolerance. Diabetes Care, 25, 1504-1510.
Lindahl, B., Nilsson, T.K., Borch-Johnsen, K., Röder, M.E., Soderberg, S., Widman, L.,
Johnson, O., Hallmanns, G., Jansson, J.H. (2009). A randomized lifestyle intervention with
5-year follow-up in subjects with impaired glucose tolerance: Pronounced short-term
impact but long-term adherence problems. Scandinavian Journal of Public Health, 37(4),
434-442.
84
Lindström, J., Eriksson, J.G.,. Valle, T.T., Aunola, S., Cepaitis, Z., Hakumaki, M.,
Hämäläinen, H., Ilanne-Parikka, P., Keinänen-Kiukaanniemi, S., Laakso, M., Louheranta,
A., Mannelin, M., Martikkala, V., Moltchanov, V., Rastas, M., Salminen, V., Sundvall, J.,
Uusitupa, M., Tuomilehto, J. (2003). Prevention of Diabetes Mellitus in Subjects with
Impaired Glucose Tolerance in the Finnish Diabetes Prevention Study: Results from a
Randomized Clinical Trial. Journal of the American Society of Nephrology; 14, S108S113.
Lindström J., Ilanne-Parikka, P., Peltonen, M., Aunola, S., Eriksson, J.G., Hemiö, K.,
Hämäläinen, H., , Härkönen, P., Keinänen-Kiukaanniemi, S., Laakso, M., Louheranta, A.,
Mannelin, M., Paturi, M., Sundvall, J., Valle T.T., Uusitupa, M., Tuomilehto, J. (2006).
Sustained reduction in the incidence of type 2 diabetes by lifestyle intervention: follow-up
of the Finnish Diabetes Prevention Study. The Lancet, 368(11), 1673-1679.
Little, P. Everitt, H., Williamson, I., Warner, G., Moore, M., Gould, C., Ferrier, K., Payne,
S. (2001). Preferences of patients for patient centred approach to consultation in primary
care: observational study. British Medical Journal, 322, 1–7
Marrero, D., Pan, Q., Barrett-Connor, E., de Groot, m., Zhang, P. Percy, C., Florez, H.,
Ackermann, R., Montez, M., Rubin, R.R. (DPPOS Research Group) (2013). Impact of
diagnosis of diabetes on health-related quality of life among high risk individuals: the
Diabetes Prevention Program outcomes study. Quality of Life Research. Retrieved from
http://download.springer.com/static/pdf/200/art%253A10.1007%252Fs11136-013-04363.pdf?auth66=1383566131_86cf77d12a0fd8fc179617aeccf4c5d4&ext=.pdf on 2th of
November 2013
Marstedt, G., Rosenbrock, R. (2010). Verhaltensprävention: Guter Wille allein reicht nicht.
in: Jan Böcken, Bernard Braun, Juliane Landmann (Hrsg.): Gesundheitsmonitor 2009 (pp.
12-37). Gütersloh, Verlag Bertelsmann Stiftung
Miron-Shatz, T., Mühlhauser, I., Bower, B., Diefenbach, M., Goldacre, B., Smith, R.S.W.,
Spiegelhalter, D., Wegwarth, O. ( 2011). Barriers to Health Information and Building
Solutions. In: G. Gigerenzer and J.A. Muir Gray (Hrsg.), Better Doctors, Better Patients,
Better Decisions – Envisioning Health Care 2020 (pp 191-212). Cambridge,
Massachusetts, London, England: The MIT Press
Mozaffarian, D., Kamineni, A. Carnethon, M., Djoussé, L., Mukamal, K.-J., Siscovick, D.
(2009). Lifestyle Risk Factors and New-Onset Diabetes Mellitus in Older Adults: The
Cardiovascular Health Study, Archives of Internal Medicine, 169(8), 798-807
85
Mozaffarian, D., Afshin, A., Benowitz, N.L., Bittner, V., Daniels, S.R., Franch, H.A.,
Jacobs, D.R. Jr., Kraus, W.E., Kris-Etherton, P.M., Krummel, D.A., Popkin, B.M., Whitsel,
L.P., Zakai, N.A. on behalf of the American Heart Association Council on Epidemiology
and Prevention, Council on Nutrition, Physical Activity and Metabolism, Council on Clinical
Cardiology, Council on Cardiovascular Disease in the Young, Council on the Kidney in
Cardiovascular Disease, Council on Peripheral Vascular Disease, and the Advocacy
Coordinating Committee. (2012). Population approaches to improve diet, physical activity,
and smoking habits: a scientific statement from the American Heart Association.
Circulation, 126, 1514 –1563.
National Institute for Health and Clinical Excellence (2011). Preventing type 2 diabetes:
population and community level interventions. NICE public health guidance 35. Retrieved
from http://www.nice.org.uk/nicemedia/live/13472/54345/54345.pdf on 15th of November
2013
National Institute for Health and Care Excellence (NICE) (2012). Preventing type 2
diabetes: risk identification and interventions for individuals at high risk. NICE public
health guidance 38.
Retrieved from http://www.nice.org.uk/nicemedia/live/13791/59951/59951.pdf on 18th of
November 2013
Nathan, D.M., Chew, E., Christophi, C.A., Davis, M.D., Fowler, S. Goldstein, B.J.,
Hamman, R.F., Hubbard, L.D. Knowler, W.C., Molitch, M.E. (Diabetes Prevention
Program Research Group) (2007). The prevalence of retinopathy in impaired glucose
tolerance and recent-onset diabetes in the Diabetes Prevention Program. Diabetic
Medicine, 24, 137–144
Orozco L.J., Buchleitner, A.M., Gimenez-Perez, G., Roque, I.F., Richter, B., Mauricio, D.
(2008). Exercise or exercise and diet for preventing type 2 diabetes mellitus. Cochrane
Database of Systematic Reviews, cited in Gillett et al., 2012
Oxford Centre for Evidence Based Medicine (2009). Levels of Evidence (March 2009).
Retrieved from http://www.cebm.net/index.aspx?o=4590 on 14th of November 2013
Oxman, A.D., Guyatt, G.H. (1991). Validation of an index of the quality of review
articles. Journal of Clinical Epidemiology, 44(11), 1271-1278.
Pan, X.R., Li, G.W., Hu, Y.H., Wang, J.X., Yang, W.Y., An, Z.X., Hu, Z.X., Lin, J., Xiao,
J.Z., Cao, H.B., Liu, P.A., Jiang, X.G., Jiang, Y.Y., Wang, J.P., Zheng, H., Zhang, H.,
Bennett, P.H., Howard, B.V. (1997). Effects of diet and exercise in preventing NIDDM in
people with impaired glucose tolerance: The Da Qing IGT and diabetes study. Diabetes
Care, 20(4), 537-44
86
Paulweber, B., Valensi, P., Lindström,J., Lalic, N.M., Greaves, J.J., McKee, M.,
Kissimova-Skarbek, K., Liatis, S., Cosson, E., Szendroedi, J., Sheppard, K.E.,
Charlesworth, K., Felton, A.-M., Hall, M., Rissannen, A., Tuomilehto, J., Schwarz, P. E.,
Roden, M. for the Writing Group,on behalf of the IMAGE Study Group (2010). A European
Evidence-Based Guideline for the Prevention of Type 2 Diabetes. Hormone and Metabolic
Research, 42, S1–S64
Penn, L.W., White, M., Oldroyd, J., Walker, M., Alberti, K.G.M.M., Mathers, J.C. (2009).
Prevention of type 2 diabetes in adults with impaired glucose tolerance: The European
Diabetes Prevention RCT in Newcastle upon Tyne, UK. BioMed Central Public Health
2009; 9, 342
Ramachandran, A., Snehalatha, C., Mary, S., Mukesh, B., Bhaskar, A. D., Vijay, V. (2006)
The Indian Diabetes Prevention Programme shows that lifestyle modification and
metformin prevent type 2 diabetes in Asian Indian subjects with impaired glucose
tolerance (IDPP-1). Diabetologia, 49(2), 289-297.
Ramachandran, A., Snehalatha, C., Satyavani, K., Sivasankari, S. Vijay, V. (2007).
Metabolic syndrome does not increase the risk of conversion of impaired glucose
tolerance to diabetes in Asian Indians: result of Indian diabetes prevention programme.
Diabetes Research and Clinical Practice, 76, 215-218
Ratner, R., Goldberg, R., Haffner, S., Marcovina, S., Orchard, T., Fowler.,S., Temprosa,
M. (2005). Impact of Intensive Lifestyle and Metformin Therapy on Cardiovascular
Disease Risk Factors in the Diabetes Prevention Program. Diabetes Care, 28(4), 888–894
Robert Koch Institut (RKI) (2005). Gesundheitsberichterstattung des Bundes. Heft 24.
Diabetes mellitus, Berlin, 14. März 2005 Retrieved from
http://www.rki.de/DE/Content/Gesundheitsmonitoring/Gesundheitsberichterstattung/GBED
ownloadsT/diabetes_mellitus.pdf?__blob=publicationFile on 6th of December 2013
Roumen, C., Corpeleijn, E., Feskens, E.J., Mensink, M., Saris, W.H., Blaak, E.E. (2008).
Impact of 3-year lifestyle intervention on postprandial glucose metabolism: the SLIM
study. Diabetic Medicine; 25(5), 597-605.
Rubin, R.R., Peyrot, M. (1999). Quality of Life and Diabetes. Diabetes/Metabolism
Research and Reviews; 15, 205-218.
Sackett, D.L., Rosenberg, W.M,. Gray, J.A., Haynes, R.B., Richardson, W.S. (1996).
Evidence based medicine: what it is and what it isn't". British Medical Journal, 312, 71-72
Santaguida, P.L., Balion, C., Hunt, D., Morrison, K., Gerstein, H., Raina, P., Booker, L.,
Yazdi, H. (2005). Diagnosis, Prognosis, and Treatment of Impaired Glucose Tolerance
and Impaired Fasting Glucose. Summary, Evidence Report/Technology Assessment No.
128. AHRQ Pub. No 05-E026-1. Rockville, MD: Agency for Healthcare Research and
Quality. Retrieved from
http://archive.ahrq.gov/downloads/pub/evidence/pdf/impglucose/impglucose.pdf on 14th of
August 2013
87
Scottish Intercollegiate Guidelines Network (2011). SIGN 50 - A guideline developer’s
handbook, First published 2008, Revised November 2011. Retrieved from www.sign.ac.uk
on 30th of October 2013
Social Code Book V, Gesetzliche Krankenversicherung, §35 b, Stand: Zuletzt geändert
durch Art. 3 G v. 7.8.2013. Retrieved from http://www.sozialgesetzbuchsgb.de/sgbv/35b.html on 30th of October 2013
Sumamo, E., Ha, C., Korownyk, C., Vandermeer, B., Dryden, D.M. (2011). Lifestyle
interventions for four conditions: type 2 diabetes, metabolic syndrome, breast cancer, and
prostate cancer. Rockville, MD, USA: Agency for Healthcare Research and Quality.
AHRQ
Technology
Assessment
Program.
2011.
Retrieved
from
http://www.cms.gov/Medicare/Coverage/DeterminationProcess/downloads/id82TA.pdf on
18th of September 2013
Tapp, R.J., Dunstan, D.W., Phillips, P. Tonkin, A., Zimmet, P.Z., Shaw, J.E. (2006).
Association between impaired glucose metabolism and quality of life: Results from the
Australian diabetes obesity and lifestyle study. Diabetes Research and Clinical Practice,
74, 154–161
Thelen, J., Kirsch, N., Hoebel, J. (2012). Gesundheit in Europa - Daten des Gesundheitsmonitorings der EU. in: Robert Koch-Institut Berlin (Hrsg.). GBE kompakt 3(6)
www.rki.de/gbe-kompakt
(Stand:
17.01.2013).
Retrieved
from
http://www.rki.de/DE/Content/Gesundheitsmonitoring/Gesundheitsberichterstattung/GBED
ownloadsK/2012_6_EUGesundheitsmonitoring.pdf;jsessionid=4BE0D8ECE925EBA15BEAD57D8D6D7645.2_cid
290?__blob=publicationFile on 15th of August 2013
Tuomilehto, J., Lindstrom, J., Eriksson, J.G., Valle, T., Aunola, S., Hämäläinen, H., IlanneParikka, P., Keinänen-Kiukaanniemi, S., Laakso, M., Louheranta, A., Rastas, M.,
Salminen, V., Uusitupa, M. (2001). Prevention of type 2 diabetes mellitus by changes in
lifestyle among subjects with impaired glucose tolerance. New England Journal of
Medicine; 344, 1343-50.
Uusitupa, M., Peltonen, M., Lindström, J., Aunola, S., Ilanne-Parikka, P., KeinänenKiukaanniemi, S., Valle, T.T., Eriksson, J.G:, Tuomilehto, J. for the Finnish Diabetes
Prevention Study Group (2009). Ten-Year Mortality and Cardiovascular Morbidity in the
Finnish Diabetes Prevention Study—Secondary Analysis of the Randomized Trial. PLoS
ONE 4(5): e5656. doi:10.1371/journal.pone.0005656
Vadstrup, E.S., Frølich, A., Perrild, H., Borg, E., Røder, M. (2011). Health-related quality
of life and self-related health in patients with type 2 diabetes: Effects of group-based
rehabilitation versus individual counseling. Health and Quality of Life Outcomes, 9, 110.
Retrieved from http://www.hqlo.com/content/9/1/110 on 16th of August 2013
88
Wein, P., Beischer, N., Harris, C., Permezel, M. (1999). A trial of simple versus intensified
dietary modification for prevention of progression to diabetes mellitus in women with
impaired glucose tolerance. Australian & New Zealand Journal of Obstetrics &
Gynaecology, 39(2), 162-166.
Wieczorek, A., Rys, P., Skrzekowska-Baran, I., Malecki, M. (2008). The Role of Surrogate
Endpoints in the Evaluation of Efficacy and Safety of Therapeutic Interventions in
Diabetes Mellitus. The Review of Diabetic Studies, 5(3), 128-135
World Cancer Research Fund/ American Institute for Cancer Research (2007). Food,
Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective.
Washington DC: AICR, 2007 Retrieved from
http://www.dietandcancerreport.org/cancer_resource_center/downloads/Second_Expert_
Report_full.pdf on 18th of December 2013
WHO (1999). Definition, diagnosis and classification of diabetes mellitus and its
complications, WHO Consultation, Retrieved from
http://www.staff.ncl.ac.uk/philip.home/who_dmg.pdf on 16th of August 2013
WHO (2003). Diet, Nutrition and the Prevention of chronic diseases. Report of a joint
WHO/FAO expert consultation, Geneva, 28 January -- 1 February 2002., Retrieved from
http://whqlibdoc.who.int/trs/who_trs_916.pdf on 18th of December 2013
WHO (2006). Definition and diagnosis of diabetes mellitus and intermediate
hyperglycemia: report of a WHO/IDF consultation. Geneva 2006. Retrieved from
http://whqlibdoc.who.int/publications/2006/9241594934_eng.pdf on 16th of August 2013
WHO (2010). 7th Global Conference on Health Promotion Track 2: Health literacy and
health behavior. Retrieved from
http://www.who.int/healthpromotion/conferences/7gchp/track2/en/index.html on 30th of
October 2013
WHO (2011). Use of Glycated Haemoglobin (HbA1c) in the Diagnosis of Diabetes
Mellitus. Abbreviated Report of a WHO Consultation. Retrieved from
http://www.who.int/diabetes/publications/report-hba1c_2011.pdf, on 26th of September
2013
WHO
Europe
(2013).
Diabetes.
Facts
and
Figures.
Retrieved
http://www.euro.who.int/en/what-we-do/health-topics/noncommunicablediseases/diabetes/facts-and-figures on 15th of August 2013
from
89
Statutory Declaration
I hereby declare that I wrote this thesis without any assistance and used only the
aids listed. Any material taken from other works, either as a quote or idea have
been indicated under “References”.
The conclusions of this thesis were reached by the author and do not necessarily
represent the views of IQWIG or its employees.
90
Appendix
Appendix 1: Application form of the O&G Index used by the IQWIG
OXMAN & GUYATT INDEX (METHODOLOGICAL QUALITY)
1.
Were the search methods used to find evidence (original research) on the primary
question(s) stated?
Yes
2.
No
Can’t tell
No
Partially
No
Was the validity of all studies referred to in the text assessed using appropriate criteria
(either in selecting studies for inclusion or in analysing the studies that are cited)?
Yes
7.
Partially
Were the criteria used for assessing the validity of the included studies reported?
Yes
6.
No
Was bias in the selection of studies avoided?
Yes
5.
Can’t tell
Were the criteria used for deciding which studies to include in the overview reported?
Yes
4.
No
Was the search for evidence reasonably comprehensive?
Yes
3.
Partially
Can’t tell
No
Were the methods used to combine the findings of the relevant studies (to reach a
conclusion) reported?
Yes
Partially
No
91
8.
Were the findings of the relevant studies combined appropriately relative to the primary
question the overview addresses?
Yes
9.
Can’t tell
Were the conclusions made by the author(s) supported by the data and/or analysis
reported in the overview?
Yes
10.
No
Partially
No
How would you rate the scientific quality of the overview?
Extensive
Major
Minor
Minimal
Flaws
flaws
flaws
flaws
1
2
3
4
5
6
7
92
Appendix 2: Project Outline
Projektskizze für Recherche
________________________
Projekt: Primärprävention des Diabetes mellitus
Typ 2 durch Lebensstil-Veränderungen
Projektcode: Z99-96-X-E0
_________________________________________________
_
Stand:
08. Mai 2013
Bearbeiterin: Martina Ehrlich und Karin Riemann-Lorenz
93
_________________________________________________
_
Hintergrund
Im Rahmen des Generalauftrags des Ressorts Gesundheitsinformation soll ein
Informationsprodukt zum Thema „Primärprävention des Diabetes mellitus Typ 2 durch
Lebensstil-Veränderungen“ erstellt werden. Außerdem soll auf Grundlage der Recherche
eine Masterarbeit (Master of Public Health) an der Fakultät Life Sciences der HAW
Hamburg, Department Gesundheitswissenschaften durch Frau Karin Riemann-Lorenz
erstellt werden.
Ziel des Projekts „Primärprävention des Diabetes mellitus Typ 2 durch LebensstilVeränderungen“ ist es, die Wirksamkeit von Lebensstilveränderungen (Änderung des
Ess- und Bewegungsverhaltens) auf die Primärprävention von Diabetes mellitus Typ 2 bei
Individuen mit erhöhtem Diabetes-Risiko nach Evidenz basierten Kriterien zu überprüfen
und darzustellen.
Die Lebensstilveränderung im Bereich Ernährung kann die Verzehrshäufigkeit bestimmter
Lebensmittelgruppen (zum Beispiel Obst- und Gemüsekonsum, Fleischkonsum,
alkoholische Getränke) betreffen, eine Änderung der Lebensmittelauswahl hinsichtlich der
Lebensmittelqualität (z.B. fettreduzierte Milchprodukte, Vollkornprodukte) umfassen und
die Nährstoffzufuhr beeinflussen (zum Beispiel Reduktion der Fettaufnahme bzw. der
Aufnahme gesättigter Fettsäuren, Erhöhung der Ballaststoffaufnahme, Reduzierung der
Energieaufnahme, Alkoholrestriktion). Supplemente und Nahrungsergänzungsmittel sowie
die Einnahme von Medikamenten zur Gewichtsreduktion sind mit Lebensstilveränderung
im Bereich Ernährung nicht gemeint. Die Lebensstilveränderung im Bereich Bewegung
bezieht sich auf eine Steigerung der körperlichen Aktivität hinsichtlich der Häufigkeit,
Dauer und/oder der Intensität. Dabei ist sowohl die Bewegung im Alltag als auch die
sportliche Betätigung gemeint.
explorative Recherche
Sichtung von Uptodate, Clinical Evidence
PubMed
(02.05.2013; limits: review; letzte 5 Jahre; Erwachsene ≥18 Jahre;
deutsch + englisch)
BioMed Central (http://www.biomedcentral.com) – Jan. bis März 2013
94
Cochrane summaries (http://summaries.cochrane.org) - Jan. bis März 2013
Wiley online library (http://onlinelibrary.wiley.com/) - Jan. bis März 2013
Springer Link (http://link.springer.com/) - Jan. bis März 2013
Testset
a. Angermayr L. Melchart D, Linde K: Multifactorial lifestyle interventions in the
primary and secondary prevention of cardiovascular disease and type 2 diabetes
mellitus--a systematic review of randomized controlled trials. Ann Behav Med.
2010 Aug;40(1):49-64. doi: 10.1007/s12160-010-9206-4
b. Michael K. Baker a, Kylie Simpson, Bradley Lloyd, Adrian E. Bauman, Maria A.
Fiatarone Singh Behavioral strategies in diabetes prevention programs: A
systematic review of randomized controlled trials. Diabetes Research and Clinical
Practice, 91(2011), S.1–12.
c. Davis N, Forges B, Wylie-Rosett J Role of obesity and lifestyle interventions in the
prevention and management of type 2 diabetes. Minerva Med. 2009
Jun;100(3):221-8
d. Dunkley AJ, Charles K, Gray LJ, Camosso-Stefinovic J, Davies MJ, Khunti K.
Effectiveness of interventions for reducing diabetes and cardiovascular disease
risk in people with metabolic syndrome: systematic review and mixed treatment
comparison meta-analysis. Diabetes Obes Metab. 2012 Jul;14(7):616-25. doi:
10.1111/j.1463-1326.2012.01571.x. Epub 2012 Feb 21
e. Gillett M, Royle P, Snaith A, Scotland G, Poobalan A, Imamura M, Black C,
Boroujerdi M, Jick S, Wyness L, McNamee P, Brennan A, Waugh N. Nonpharmacological interventions to reduce the risk of diabetes in people with
impaired glucose regulation: a systematic review and economic evaluation. Health
Technol Assess. 2012 Aug;16(33):1-236, iii-iv. doi: 10.3310/hta16330
f.
Horton ES Effects of lifestyle changes to reduce risks of diabetes and associated
cardiovascular risks: results from large scale efficacy trials. Obesity (Silver Spring).
2009 Dec;17 Suppl 3:S43-8. doi: 10.1038/oby.2009.388
g. Roumen C, Blaak EE, Corpeleijn E. Lifestyle intervention for prevention of
diabetes: determinants of success for future implementation. Nutr Rev. 2009
Mar;67(3):132-46. doi: 10.1111/j.1753-4887.2009.00181.x
h. C. Sanz, J.-F.Gautier, H. Hanaire Physical exercise for the prevention and
treatment of type 2 diabetes Doi : 10.1016/j.diabet.2010.06.001
i.
Thomas GN, Jiang CQ, Taheri S, Xiao ZH, Tomlinson B, Cheung BM, Lam TH,
Barnett AH, Cheng KK A systematic review of lifestyle modification and glucose
intolerance in the prevention of type 2 diabetes. Curr Diabetes Rev. 2010
Nov;6(6):378-87.
95
j.
Tuomilehto, Jaakko Nonpharmacologic Therapy and Exercise in the Prevention of
Type 2 Diabetes Care 32 (Suppl. 2):S189–S193, 2009
k. Yamaoka, Kazue and Toshiro Tango 2012. Effects of lifestyle modification on
metabolic syndrome: a systematic review and meta-analysis. BMC Medicine 2012,
10:138 http://www.biomedcentral.com/1741-7015/10/138.
l.
Yoon, Uzung, Kwok, Lai Lai. and Magkidis, Athanasios. Efficacy of lifestyle
interventions in reducing diabetes incidence in patients with impaired glucose
tolerance: A systematic review of randomized controlled trials. Metabolism Clinical and Experimental Volume 62, Issue 2 , Pages 303-314, February 2013
m. Yuen, Agnes, Sugeng, Yulia, Weiland, Tracey J. and George A. Jelinek, Lifestyle
and medication interventions for the prevention or delay of type 2 diabetes mellitus
in prediabetes: a systematic review of randomised controlled trials. Australian and
New Zealand Journal of Public Health, 34(2), pages.172–178, April 2010
DOI: 10.1111/j.1753-6405.2010.00503.x
PICO
Einschlusskriterien
Intervention
Adults (≥ 18 years) without diagnosis of diabetes mellitus
before entering study and with elevated risk of diabetes
mellitus type 2 defined by study (e.g. Impaired Glucose
Tolerance, elevated fasting blood glucose, obesity)
lifestyle intervention including change in diet and/or
physical activity
Control
General advice/usual care, pharmacological treatment,
no treatment
E3
Outcome
Diagnosis of DM 2, period until diagnosis of DM 2,overall
mortality, cardiovascular morbidity and mortality,
microvascular diseases, quality of life or clinical
parameters (reduction in blood glucose, blood pressure,
BMI)
E4
Studientyp
Meta-analysis or systematic review of controlled studies
(CT`s), HTA
E5
Recherche
Die der Arbeit zugrunde liegende Recherche erfolgte
2009 oder später.
E6
Publikationssprache
Englisch oder Deutsch
E7
Veröffentlichung
Volltext-Publikation beschaffbar/vorhanden
E8
Oxman & Guyatt
Qualität des Reviews nach O & G ≥ 5
E9
Scope
Thema/Fragestellung relevant
E10
Population
E1
E2
Informationsbeschaffung
96
gesuchte Studientypen
-
Systematische Reviews aus kontrollierten Studien
-
HTAs
Bibliografische Literaturrecherche
Die systematische Literaturrecherche nach relevanten systematischen Übersichten soll in
folgenden Quellen durchgeführt werden:
-
Medline (Ovid)
PubMed
Cochrane Database of Systematic Reviews (Cochrane Reviews)
Database of Abstracts of Reviews of Effects
Health Technology Assessment Database (Technology Assessments)
Autorenanfragen sind nicht vorgesehen.
97
Appendix 3: Exclusion criteria for studies in the full text screening process –
results of consenting process
Study
Allende-Vigo, M. Z. Diabetes Mellitus Prevention. Am J Ther 2011
Angermayr, Lucia; Melchart, Dieter; Linde, Klaus Multifactorial lifestyle
interventions in the primary and secondary prevention of cardiovascular
disease and type 2 diabetes mellitus—a systematic review of randomized
controlled trials. Annals of Behavioral Medicine 2010; 40(1):49- 64
Baker, Michael K.; Simpson, Kylie; Lloyd, Bradley; Bauman, Adrian E.;
Singh, Maria A. Fiatarone Behavioral strategies in diabetes prevention
programs: a systematic review of randomized controlled trials. Diabetes
Research & Clinical Practice 2011; 91(1):1-12
Bonfioli, E.; Berti, L.; Goss, C.; Muraro, F.; Burti, L. Health promotion lifestyle
interventions for weight management in psychosis: a systematic review and
meta-analysis of randomized controlled trials. BMC Psychiatry 2012; 12():78
Brown, T.; Avenell, A.; Edmunds, L. D.; Moore, H.; Whittaker, V.; Avery, L.;
Summerbell, C. Systematic review of long-term lifestyle interventions to
prevent weight gain and morbidity in adults. Obesity Reviews 2009;
10(6):627-38
Cardona-Morrell, Magnolia; Rychetnik, Lucie; Morrell, Stephen L.; Espinel,
Paola T.; Bauman, Adrian Reduction of diabetes risk in routine clinical
practice: are physical activity and nutrition interventions feasible and are the
outcomes from reference trials replicable? A systematic review and metaanalysis. BMC Public Health 2010; 10():653
Conn, V. S.; Hafdahl, A. R.; Brown, L. M. Meta-analysis of quality-of-life
outcomes from physical activity interventions (Structured abstract). Nursing
Research 2009; 58(3):175-183
Dombrowski, S. U.; Avenell, A.; Sniehott, F. F. Behavioural interventions for
obese adults with additional risk factors for morbidity: aystematic review of
effects on behaviour, weight and disease risk factors (Structured abstract).
Obesity Facts 2010; 3(6):377-396
Dyson, P. A. The therapeutics of lifestyle management on obesity. Diabetes,
Obesity & Metabolism 2010; 12(11):941-6
Esposito, Katherine; Maiorino, Maria Ida; Ceriello, Antonio; Giugliano, Dario
Prevention and control of type 2 diabetes by Mediterranean diet: a
systematic review. Diabetes Research & Clinical Practice 2010; 89(2):97102
Esposito, K.; Kastorini, C. M.; Panagiotakos, D. B.; Giugliano, D.
Mediterranean diet and weight loss: meta-analysis of randomized controlled
trials (Structured abstract). Metabolic Syndrome and Related Disorders
2011; 9(1):1-12
Gillett M, Royle P, Snaith A, Scotland G, Poobalan A, Imamura M, Black C,
et al. Non-pharmacological interventions to reduce the risk of diabetes in
people with impaired glucose regulation: a systematic review and economic
evaluation. Health Technol Assess 2012; 16(33): 1-236.
Hopper, Ingrid; Billah, Baki; Skiba, Marina; Krum, Henry Prevention of
diabetes and reduction in major cardiovascular events in studies of subjects
with prediabetes: meta-analysis of randomised controlled clinical trials.
European Journal of Cardiovascular Prevention & Rehabilitation 2011;
18(6):813-23
Hu, Tian; Mills, Katherine T; .; Yao, Lu; Demanelis, Kathryn; Eloustaz,
Mohamed; Yancy, William S., Jr.; Kelly, Tanika N.; He, Jiang; Bazzano,
Lydia A. Effects of low-carbohydrate diets versus low-fat diets on metabolic
risk factors: a meta-analysis of randomized controlled clinical trials.
American Journal of Epidemiology 2012; 176 Suppl 7():S44-54
Jackson, Lindsey Translating the Diabetes Prevention Program into practice:
a review of community interventions. Diabetes Educator 2009; 35(2):309-20
Reason for
exclusion
E5
E1
E9
E1
E6
E10
E1
E6
E10
E1
E10
E9
E9
E10
E5
98
Johnson, M.; Jones, R.; Freeman, C.; Woods, H. B.; Gillett, M.; Goyder, E.;
Payne, N. Can diabetes prevention programmes be translated effectively
into real-world settings and still deliver improved outcomes? A synthesis of
evidence. Diabetic Medicine 2013; 30(1):3-15
Kastorini, Christina-Maria; Milionis, Haralampos J.; Esposito, Katherine;
Giugliano, Dario; Goudevenos, John A.; Panagiotakos, Demosthenes B. The
effect of Mediterranean diet on metabolic syndrome and its components: a
meta-analysis of 50 studies and 534,906 individuals. Journal of the
American College of Cardiology 2011; 57(11):1299-313
Katzmarzyk, P. T.; Lear, S. A. Physical activity for obese individuals: a
systematic review of effects on chronic disease risk factors (Structured
abstract). Obesity Reviews 2012; 13(2):95-105
Koivula, R. W.; Tornberg, A. B.; Franks, P. W. Exercise and diabetes-related
cardiovascular disease: systematic review of published evidence from
observational studies and clinical trials. Current Diabetes Reports 2013;
13(3):372-80
Leao, Leila Sicupi; ra Carneiro de Souza; de Moraes, Milena Miranda; de
Carvalho, Giulia Xavier; Koifman, Rosalina Jorge Nutritional interventions in
metabolic syndrome: a systematic review. Arquivos Brasileiros de
Cardiologia 2011; 97(3):260-5
Leblanc, Erin S.; O; '; Connor, Elizabeth; Whitlock, Evelyn P.; Patnode,
Carrie D.; Kapka, Tanya Effectiveness of primary care-relevant treatments
for obesity in adults: a systematic evidence review for the U.S. Preventive
Services Task Force. Annals of Internal Medicine 2011; 155(7):434-47
Liu, Zhao-Min; Chen, Yu-Ming; Ho, Suzanne C. Effects of soy intake on
glycemic control: a meta-analysis of randomized controlled trials. American
Journal of Clinical Nutrition 2011; 93(5):1092-101
Lukacova-Zib, Ivana; Gopalakrishnan, Geetha Therapeutic options for the
prevention of type 2 diabetes mellitus in the metabolic syndrome. Mount
Sinai Journal of Medicine 2010; 77(5):524-32
Osei-Assibey, G and and C Boachie Dietary interventions for weight loss
and cardiovascular risk reduction in people of African ancestry (blacks): a
systematic review Public Health Nutrition: 15(1), 110–115
Pattyn, N.; Cornelissen, V. A.; Eshghi, S. R.; Vanhees, L. The effect of
exercise on the cardiovascular risk factors constituting the metabolic
syndrome: a meta-analysis of controlled trials. Sports Medicine 2013;
43(2):121-33
Rawal, Lal B.; Tapp, Robyn J.; Williams, Emily D.; Chan, Carina; Yasin,
Shajahan; Oldenburg, Brian Prevention of type 2 diabetes and its
complications in developing countries: a review. International Journal of
Behavioral Medicine 2012; 19(2):121-33
Schwingshackl, L.; Strasser, B.; Hoffmann, G. Effects of monounsaturated
fatty acids on glycaemic control in patients with abnormal glucose
metabolism: a systematic review and metaanalysis. Annals of Nutrition &
Metabolism 2011; 58(4):290-6
Shirani, F.; Salehi-Abargouei, A.; Azadbakht, L. Effects of Dietary
Approaches to Stop Hypertension (DASH) diet on some risk for developing
type 2 diabetes: A systematic review and meta-analysis on controlled clinical
trials. Nutrition 2013;
Shrestha, Prabha; Ghimire, Laxmi A review about the effect of life style
modification on diabetes and quality of life. Global Journal of Health Science
2012; 4(6):185-90
Sievenpiper, J. L.; Kendall, C. W.; Esfahani, A.; Wong, J. M.; Carleton, A. J.;
Jiang, H. Y.; Bazinet, R. P.; Vidgen, E.; Jenkins, D. J. Effect of non-oil-seed
pulses on glycaemic control: a systematic review and meta-analysis of
randomised controlled experimental trials in people with and without
diabetes (Structured abstract). Diabetologia 2009; 52(8):1479-1495
Steyn, Nelia P.; Lambert, Estelle V.; Tabana, Hanani Conference on
"Multidisciplinary approaches to nutritional problems". Symposium on
"Diabetes and health". Nutrition interventions for the prevention of type 2
E5
E10
E10
E1
E10
E10
E10
E5
E1
E10
E10
E10
E1
E5
E6
E6
99
diabetes. Proceedings of the Nutrition Society 2009; 68(1):55-70
Strasser, B.; Siebert, U.; Schobersberger, W. Resistance training in the
treatment of the metabolic syndrome: a systematic review and meta-analysis
of the effect of resistance training on metabolic clustering in patients with
abnormal glucose metabolism (Provisional abstract). Sports Medicine 2010;
40(5):397-415
Thomas, G. Neil; Jiang, Chao Q.; Taheri, Shahrad; Xiao, Zheng H.;
Tomlinson, Brian; Cheung, Bernard M. Y.; Lam, Tai H.; Barnett, Anthony H.;
Cheng, Kar K. A systematic review of lifestyle modification and glucose
intolerance in the prevention of type 2 diabetes. Current Diabetes Reviews
2010; 6(6):378-87
Thompson, Elizabeth; Berry, Diane; Nasir, Laura Weight management in
African-Americans using church-based community interventions to prevent
type 2 diabetes and cardiovascular disease. Journal of National Black
Nurses Association 2009; 20(1):59-65
Tourlouki, Eleni; Matalas, Antonia-Leda; Panagiotakos, Demosthenes B.
Dietary habits and cardiovascular disease risk in middle-aged and elderly
populations: a review of evidence. Clinical Interventions In Aging 2009;
4():319-30
Tschentscher, M.; Niederseer, D.; Niebauer, J. Health benefits of nordic
walking: a systematic review. American Journal of Preventive Medicine
2013; 44(1):76-84
Whittemore, R. A systematic review of the translational research on the
Diabetes Prevention Program (Provisional abstract). Translational
Behavioral Medicine 2011; 1(3):480-491
Wolfram, Taylor; Ismail-Beigi, Faramarz Efficacy of high-fiber diets in the
management of type 2
diabetes mellitus. Endocrine Practice 2011; 17(1):132-42
Wycherley, T. P.; Moran, L. J.; Clifton, P. M.; Noakes, M.; Brinkworth, G. D.
Effects of energyrestricted high-protein, low-fat compared with standardprotein, low-fat diets: a meta-analysis of randomized controlled trials
(Provisional abstract). Database of Abstracts of Reviews of Effects 2012;
(2):1281-1298
Yamaoka, K.; Tango, T. Efficacy of lifestyle education in preventing type 2
diabetes: an updated version (Provisional abstract). Salud (i) Ciencia 2009;
17(1):29-33
Yamaoka, Kazue; Tango, Toshiro Effects of lifestyle modification on
metabolic syndrome: a systematic review and meta-analysis. BMC Medicine
2012; 10():138
Yeh, G. Y.; Wang, C.; Wayne, P. M.; Phillips, R. Tai chi exercise for patients
with cardiovascular conditions and risk factors: a systematic review
(Structured abstract). Journal of Cardiopulmonary Rehabilitation and
Prevention 2009; 29(3):152-160
Yoon, Uzung, Lai Lai Kwok, Athanasios Magkidis Efficacy of lifestyle
interventions in reducing diabetes incidence in patients with impaired
glucose tolerance: A systematic review of randomized controlled trials,
Metabolism Clinical and Experimental 62 (2013) 303-314
Yuen, A.; Sugeng, Y.; Weiland, T. J.; Jelinek, G. A. Lifestyle and medication
interventions for the prevention or delay of type 2 diabetes mellitus in
prediabetes: a systematic review of randomised controlled trials. Aust N Z J
Public Health 2010; 34(2):172-8
E6
E9
E6
E5
E10
E5
E1
E10
E7
E10
E6
E9
E6
100
Appendix 4: Evidence grading systems (ADA and SIGN)
Sign Criteria
LEVELS OF EVIDENCE
High quality meta-analyses, systematic reviews of RCTs, or RCTs with a very low
1++
risk of bias
Well conducted meta-analyses, systematic reviews, or RCTs with a low risk of bias
1+
Meta-analyses, systematic reviews, or RCTs with a high risk of bias
1
High quality systematic reviews of case control or cohort studies
2++
High quality case control or cohort studies with a very low risk of confounding or bias
and a high probability that the relationship is causal
Well conducted case control or cohort studies with a low risk of confounding or bias
2+
and a moderate probability that the relationship is causal
Case control or cohort studies with a high risk of confounding or bias and a significant
2risk that the relationship is not causal
Non-analytic studies, eg case reports, case series
3
Expert opinion
4
GRADES OF RECOMMENDATION
Note: The grade of recommendation relates to the strength of the evidence on which the
recommendation is based. It does not reflect the clinical importance of the recommendation
A
• At least one meta-analysis, systematic review, or RCT rated as 1++, and directly
applicable to the target population; or
• A body of evidence consisting principally of studies rated as 1+, directly
applicable to the target population, and demonstrating overall consistency of
results
B
• A body of evidence including studies rated as 2++, directly applicable to the
target population, and demonstrating overall consistency of results; or
• Extrapolated evidence from studies rated as 1++ or 1+
C
• A body of evidence including studies rated as 2+, directly applicable to the
target population and demonstrating overall consistency of results; or
• Extrapolated evidence from studies rated as 2++
D
• Evidence level 3 or 4; or
• Extrapolated evidence from studies rated as 2+
Source: Scottish Intercollegiate Guidelines Network (2011). SIGN 50 - A guideline developer’s
handbook, First published 2008, Revised November 2011, download from www.sign.ac.uk
101
ADA evidence grading system for clinical practice recommendations
A
Clear evidence from well-conducted, generalizable RCTs that are adequately powered,
including:
• Evidence from a well-conducted multicenter trial
• Evidence from a meta-analysis that incorporated quality ratings in the analysis
Compelling non-experimental evidence, i.e., “all or none” rule developed by the Centre for
Evidence-Based Medicine at the University of Oxford
B
C
D
Supportive evidence from well-conducted RCTs that are adequately powered,
including:
• Evidence from a well-conducted trial at one or more institutions
• Evidence from a meta-analysis that incorporated quality ratings in the analysis
Supportive evidence from well-conducted cohort studies
• Evidence from a well-conducted prospective cohort study or registry
• Evidence from a well-conducted meta-analysis of cohort studies
Supportive evidence from a well-conducted case-control study
Supportive evidence from poorly controlled or uncontrolled studies
• Evidence from randomized clinical trials with one or more major or three or more
minor methodological flaws that could invalidate the results
• Evidence from observational studies with high potential for bias (such as case
series with comparison with historical controls)
• Evidence from case series or case reports
Conflicting evidence with the weight of evidence supporting the recommendation
Expert consensus or clinical experience
Source: ADA (2013). Standards of Medical Care in Diabetes- 2013. Position Statement.
Diabetes Care, Volume 36, Supplement 1, January 2013 S.11-S.66
102
Appendix 5: Oxford Centre for Evidence Based Medicine: Levels of Evidence
(March 2009) download from http://www.cebm.net/index.aspx?o=4590
103
104